Podcast appearances and mentions of Daniel Goldstein

Happy holidays! We'll be sharing snippets from Latent Space LIVE! through the break bringing you the best of 2024! We want to express our deepest appreciation to event sponsors AWS, Daylight Computer, Thoth.ai, StrongCompute, Notable Capital, and most of all all our LS supporters who helped fund the gorgeous venue and A/V production!For NeurIPS last year we did our standard conference podcast coverage interviewing selected papers (that we have now also done for ICLR and ICML), however we felt that we could be doing more to help AI Engineers 1) get more industry-relevant content, and 2) recap 2024 year in review from experts. As a result, we organized the first Latent Space LIVE!, our first in person miniconference, at NeurIPS 2024 in Vancouver.Of perennial interest, particularly at academic conferences, is scaled-up architecture research as people hunt for the next Attention Is All You Need. We have many names for them: “efficient models”, “retentive networks”, “subquadratic attention” or “linear attention” but some of them don't even have any lineage with attention - one of the best papers of this NeurIPS was Sepp Hochreiter's xLSTM, which has a particularly poetic significance as one of the creators of the LSTM returning to update and challenge the OG language model architecture:So, for lack of a better term, we decided to call this segment “the State of Post-Transformers” and fortunately everyone rolled with it.We are fortunate to have two powerful friends of the pod to give us an update here:* Together AI: with CEO Vipul Ved Prakash and CTO Ce Zhang joining us to talk about how they are building Together together as a quote unquote full stack AI startup, from the lowest level kernel and systems programming to the highest level mathematical abstractions driving new model architectures and inference algorithms, with notable industry contributions from RedPajama v2, Flash Attention 3, Mamba 2, Mixture of Agents, BASED, Sequoia, Evo, Dragonfly, Dan Fu's ThunderKittens and many more research projects this year* Recursal AI: with CEO Eugene Cheah who has helped lead the independent RWKV project while also running Featherless AI. This year, the team has shipped RWKV v5, codenamed Eagle, to 1.5 billion Windows 10 and Windows 11 machines worldwide, to support Microsoft's on-device, energy-usage-sensitive Windows Copilot usecases, and has launched the first updates on RWKV v6, codenamed Finch and GoldFinch. On the morning of Latent Space Live, they also announced QRWKV6, a Qwen 32B model modified with RWKV linear attention layers. We were looking to host a debate between our speakers, but given that both of them were working on post-transformers alternativesFull Talk on YoutubePlease like and subscribe!LinksAll the models and papers they picked:* Earlier Cited Work* Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention* Hungry hungry hippos: Towards language modeling with state space models* Hyena hierarchy: Towards larger convolutional language models* Mamba: Linear-Time Sequence Modeling with Selective State Spaces* S4: Efficiently Modeling Long Sequences with Structured State Spaces* Just Read Twice (Arora et al)* Recurrent large language models that compete with Transformers in language modeling perplexity are emerging at a rapid rate (e.g., Mamba, RWKV). Excitingly, these architectures use a constant amount of memory during inference. However, due to the limited memory, recurrent LMs cannot recall and use all the information in long contexts leading to brittle in-context learning (ICL) quality. A key challenge for efficient LMs is selecting what information to store versus discard. In this work, we observe the order in which information is shown to the LM impacts the selection difficulty. * To formalize this, we show that the hardness of information recall reduces to the hardness of a problem called set disjointness (SD), a quintessential problem in communication complexity that requires a streaming algorithm (e.g., recurrent model) to decide whether inputted sets are disjoint. We empirically and theoretically show that the recurrent memory required to solve SD changes with set order, i.e., whether the smaller set appears first in-context. * Our analysis suggests, to mitigate the reliance on data order, we can put information in the right order in-context or process prompts non-causally. Towards that end, we propose: (1) JRT-Prompt, where context gets repeated multiple times in the prompt, effectively showing the model all data orders. This gives 11.0±1.3 points of improvement, averaged across 16 recurrent LMs and the 6 ICL tasks, with 11.9× higher throughput than FlashAttention-2 for generation prefill (length 32k, batch size 16, NVidia H100). We then propose (2) JRT-RNN, which uses non-causal prefix-linear-attention to process prompts and provides 99% of Transformer quality at 360M params., 30B tokens and 96% at 1.3B params., 50B tokens on average across the tasks, with 19.2× higher throughput for prefill than FA2.* Jamba: A 52B Hybrid Transformer-Mamba Language Model* We present Jamba, a new base large language model based on a novel hybrid Transformer-Mamba mixture-of-experts (MoE) architecture. * Specifically, Jamba interleaves blocks of Transformer and Mamba layers, enjoying the benefits of both model families. MoE is added in some of these layers to increase model capacity while keeping active parameter usage manageable. * This flexible architecture allows resource- and objective-specific configurations. In the particular configuration we have implemented, we end up with a powerful model that fits in a single 80GB GPU.* Built at large scale, Jamba provides high throughput and small memory footprint compared to vanilla Transformers, and at the same time state-of-the-art performance on standard language model benchmarks and long-context evaluations. Remarkably, the model presents strong results for up to 256K tokens context length. * We study various architectural decisions, such as how to combine Transformer and Mamba layers, and how to mix experts, and show that some of them are crucial in large scale modeling. We also describe several interesting properties of these architectures which the training and evaluation of Jamba have revealed, and plan to release checkpoints from various ablation runs, to encourage further exploration of this novel architecture. We make the weights of our implementation of Jamba publicly available under a permissive license.* SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformers* We introduce Sana, a text-to-image framework that can efficiently generate images up to 4096×4096 resolution. Sana can synthesize high-resolution, high-quality images with strong text-image alignment at a remarkably fast speed, deployable on laptop GPU. Core designs include: * (1) Deep compression autoencoder: unlike traditional AEs, which compress images only 8×, we trained an AE that can compress images 32×, effectively reducing the number of latent tokens. * (2) Linear DiT: we replace all vanilla attention in DiT with linear attention, which is more efficient at high resolutions without sacrificing quality. * (3) Decoder-only text encoder: we replaced T5 with modern decoder-only small LLM as the text encoder and designed complex human instruction with in-context learning to enhance the image-text alignment. * (4) Efficient training and sampling: we propose Flow-DPM-Solver to reduce sampling steps, with efficient caption labeling and selection to accelerate convergence. * As a result, Sana-0.6B is very competitive with modern giant diffusion model (e.g. Flux-12B), being 20 times smaller and 100+ times faster in measured throughput. Moreover, Sana-0.6B can be deployed on a 16GB laptop GPU, taking less than 1 second to generate a 1024×1024 resolution image. Sana enables content creation at low cost. * RWKV: Reinventing RNNs for the Transformer Era* Transformers have revolutionized almost all natural language processing (NLP) tasks but suffer from memory and computational complexity that scales quadratically with sequence length. In contrast, recurrent neural networks (RNNs) exhibit linear scaling in memory and computational requirements but struggle to match the same performance as Transformers due to limitations in parallelization and scalability. * We propose a novel model architecture, Receptance Weighted Key Value (RWKV), that combines the efficient parallelizable training of transformers with the efficient inference of RNNs.* Our approach leverages a linear attention mechanism and allows us to formulate the model as either a Transformer or an RNN, thus parallelizing computations during training and maintains constant computational and memory complexity during inference. * We scale our models as large as 14 billion parameters, by far the largest dense RNN ever trained, and find RWKV performs on par with similarly sized Transformers, suggesting future work can leverage this architecture to create more efficient models. This work presents a significant step towards reconciling trade-offs between computational efficiency and model performance in sequence processing tasks.* LoLCATs: On Low-Rank Linearizing of Large Language Models* Recent works show we can linearize large language models (LLMs) -- swapping the quadratic attentions of popular Transformer-based LLMs with subquadratic analogs, such as linear attention -- avoiding the expensive pretraining costs. However, linearizing LLMs often significantly degrades model quality, still requires training over billions of tokens, and remains limited to smaller 1.3B to 7B LLMs. * We thus propose Low-rank Linear Conversion via Attention Transfer (LoLCATs), a simple two-step method that improves LLM linearizing quality with orders of magnitudes less memory and compute. * We base these steps on two findings. * First, we can replace an LLM's softmax attentions with closely-approximating linear attentions, simply by training the linear attentions to match their softmax counterparts with an output MSE loss ("attention transfer").* Then, this enables adjusting for approximation errors and recovering LLM quality simply with low-rank adaptation (LoRA). * LoLCATs significantly improves linearizing quality, training efficiency, and scalability. We significantly reduce the linearizing quality gap and produce state-of-the-art subquadratic LLMs from Llama 3 8B and Mistral 7B v0.1, leading to 20+ points of improvement on 5-shot MMLU. * Furthermore, LoLCATs does so with only 0.2% of past methods' model parameters and 0.4% of their training tokens. * Finally, we apply LoLCATs to create the first linearized 70B and 405B LLMs (50x larger than prior work). * When compared with prior approaches under the same compute budgets, LoLCATs significantly improves linearizing quality, closing the gap between linearized and original Llama 3.1 70B and 405B LLMs by 77.8% and 78.1% on 5-shot MMLU.Timestamps* [00:02:27] Intros* [00:03:16] Why Scale Context Lengths? or work on Efficient Models* [00:06:07] The Story of SSMs* [00:09:33] Idea 1: Approximation -> Principled Modeling* [00:12:14] Idea 3: Selection* [00:15:07] Just Read Twice* [00:16:51] Idea 4: Test Time Compute* [00:17:32] Idea 2: Hardware & Kernel Support* [00:19:49] RWKV vs SSMs* [00:24:24] RWKV Arch* [00:26:15] QWRKWv6 launch* [00:30:00] What's next* [00:33:21] Hot Takes - does anyone really need long context?Transcript[00:00:00] AI Charlie: We're back at Latent Space Live, our first mini conference held at NeurIPS 2024 in Vancouver. This is Charlie, your AI co host. As a special treat this week, we're recapping the best of 2024 going domain by domain. We sent out a survey to the over 900 of you who told us what you wanted, and then invited the best speakers in the Latent Space Network to cover each field.[00:00:24] AI Charlie: 200 of you joined us in person throughout the day, with over 2200 watching live online. Thanks Our next keynote covers the State of Transformers alternative architectures, with a special joint presentation with Dan Fu of Together AI and Eugene Chia of Recursal AI and Featherless AI. We've featured both Together and Recursal on the pod before, with CEO Veepal Vedprakash introducing them.[00:00:49] AI Charlie: And CTO CE Zhang joining us to talk about how they are building together together as a quote unquote full stack AI startup from the lowest level kernel and systems [00:01:00] programming to the highest level mathematical abstractions driving new model architectures and inference algorithms with notable industry contributions from Red Pajama V2, Flash Attention 3, Mamba 2, Mixture of Agents.[00:01:15] AI Charlie: Based, Sequoia, Evo, Dragonfly, Danfoo's Thunder Kittens, and many more research projects this year. As for Recursal and Featherless, we were the first podcast to feature RWKV last year, and this year the team has shipped RWKV v5, codenamed Eagle, to 1. 5 billion Windows 10 and Windows 11 machines worldwide to support Microsoft's on device, end Energy Usage Sensitive Windows Copilot Use Cases and has launched the first updates on RWKV v6, codenamed Finch and Goldfinch.[00:01:53] AI Charlie: On the morning of Latent Space Live, they also announced QRdata UKv6, a QEN32B model [00:02:00] modified with RDWKV linear attention layers. Eugene has also written the most single most popular guest post on the Latent Space blog this year. Yes, we do take guest posts on what he has discovered about the H100 GPU inference NeoCloud market since the successful launch of Featherless AI this year.[00:02:20] AI Charlie: As always, don't forget to check the show notes for the YouTube link to their talk as well as their slides. Watch out and take care.[00:02:27] Intros[00:02:27] Dan Fu: Yeah, so thanks so much for having us. So this is going to be a little bit of a two part presentation. My name is Dan. I'm at Together AI, and I'll be joining UCSD as faculty in about a year. And Eugene, you want to introduce yourself?[00:02:46] Eugene Cheah: Eugene, I lead the art activity team, and I, I'm CEO of Featherless, and we both work on this new post transformer architecture space.[00:02:55] Dan Fu: Yeah, so yeah, so today we're really excited to talk to you a little bit [00:03:00] about that. So first I'm going to give a broad overview of kind of the last few years of progress in non post transformer architectures. And then afterwards Eugene will tell us a little bit about the latest and the greatest and the latest frontier models in this space.[00:03:16] Why Scale Context Lengths? or work on Efficient Models[00:03:16] Dan Fu: So, the story starts with Scaling. So this is probably a figure or something like this that you've seen very recently. Over the last five to six years, we've seen models really scale up in parameter size, and that's brought with it a bunch of new capabilities, like the ability to talk to you and tell you sometimes how to use your Colab screens.[00:03:35] Dan Fu: But another place where we've seen scaling especially recently is scaling in context length. So this can mean Having more text inputs for your models, but it can also mean things like taking a lot of visual token inputs image inputs to your models or generating lots of outputs. And one thing that's been really exciting over the last few months or so is that we're, we're seeing scaling, not only during training time, but also [00:04:00] during test time.[00:04:00] Dan Fu: So this is one of the, the, this is the iconic image from the OpenAI 01 release. Not only are we starting to scale train time compute, but we're also starting to scale test time compute. Now if you're familiar with our attention and our transformer architectures today, this graph on the right might look a little bit scary.[00:04:19] Dan Fu: And one of the reasons is that the implications are a little bit Interesting. So what does it mean if we want to continue having smarter and smarter models? Do we just need to start building bigger, bigger data centers, spending more flops? Is this this little Dolly 3, we need more flops, guys? Is this going to be the future of all of AI?[00:04:39] Dan Fu: Or is there a better way, another path forward? Maybe we can get the same capabilities that we've gotten used to, But for a lot less compute, a lot less flops. And one of the things that we're going to talk about today is specifically looking at that core attention operator in some of these models.[00:04:57] Dan Fu: And the reason is that so this is just some, some [00:05:00] basic you know, scaling curves, but attention has compute that scales quadratically in the context length. So that means that if you're doing something like test time compute and you want to spend a bunch of tokens thinking about what comes next, the longer that that goes the, the, the more tokens you spend on that, that compute grows quadratically in that.[00:05:19] Dan Fu: One of the questions that we're interested in is, can we take that basic sequence model, that basic sequence primitive at the bottom, and get it to scale better? Can we scale in, let's say, n to the 3 halves or n log n? So in, in the first part of the talk, so we just went over the introduction. What I'm gonna do over the next few slides is just talk about some of the key advances and ideas that have shown over the past few years since maybe early 2020 to, to now that shown promise that this might actually be possible.[00:05:48] Dan Fu: That you can actually get potentially the same quality that we want while scale, while scaling better. So to do that, we're and, and basically the, the story that we're gonna look is we're gonna start to see [00:06:00] how. So this is a basic graph of just the past couple years of progress of perplexity where that blue line, that dotted blue line, is attention.[00:06:07] The Story of SSMs[00:06:07] Dan Fu: It's your basic transformer, full dense attention. And then the dots coming down are some of the methods that you'll see in this presentation today. We're going to turn the clock back all the way to 2020. So this, this, this question of can we make attention subquadratic? Basically, as soon as we said attention is all you need, People started asking this question.[00:06:28] Dan Fu: So we have this quadratic attention operator. Can we do better? I'll briefly talk about why attention is quadratic. And the basic thing that happens, if you're not familiar, is that you have these inputs, these keys and queries. And what you do in this attention matrix, this S matrix over here, is that you're using, you're comparing every token in your input to every other token.[00:06:49] Dan Fu: So when I try to do something like upload a whole book to Gemini, what happens beyond the Maybe not Gemini, because we don't necessarily know what architecture is. But let's say we upload it to LLAMA, what happens beyond [00:07:00] the scenes, behind the scenes, is that it's going to take every single word in that book and compare it to every other word.[00:07:05] Dan Fu: And this has been a really, it's, it's led to some pretty impressive things. But it's kind of a brute forcing of the way that you would try to interpret a interpret something. And what attention does in particular is the, and then what attention, sorry, don't want to. Okay, no, no laser pointer. What, what attention does afterwards is that instead of always operating in this quadratic thing, it takes a row wise softmax over this matrix, and then multiplies it by this values matrix.[00:07:32] Dan Fu: So, one of the key points to notice is that the output size is always going to be the same as the inputs, at least in standard self attention. So one of the first things that folks tried to do around 2020 is this thing called linear attention, which is just, just noticing that if we take out this softmax from here, if we take out this non linearity in the middle of the attention operation, and then if you compute the keys and the values operation first, you actually never hit this quadratic bottleneck.[00:07:57] Dan Fu: So that, that's potentially a way [00:08:00] to get a lot more computationally efficient. And there are various ways to do this by basically using feature maps or try to approximate this overall attention computation. But some of this work sort of started to hit a wall in 2020. And the basic challenges were, were two.[00:08:16] Dan Fu: So one was quality. It was back then, it was kind of hard to, to get good quality with these linear attention operators. The other one was actually hardware efficiency. So these, this feature map that was just shown by a simplify simplify here. Actually ends up being quite computationally expensive if you just implement it naively.[00:08:34] Dan Fu: So you started having these operators that not only were you sure, you're not really sure if they have the same quality, but also they're actually just wall clock slower. So you kind of end up getting the worst of both worlds. So this was the the stage. So that kind of sets the stage for four years ago.[00:08:49] Dan Fu: Keep this in mind because linear attention is actually going to come back in a few years once we have a better understanding. But one of the works that started kicking off this, this [00:09:00] mini revolution in post transformer architectures was this idea called states based model. So here the seminal work is, is one about our work queue in 2022.[00:09:09] Dan Fu: And this, this piece of work really brought together a few ideas from, from some long running research research lines of work. The first one was, and this is really one of the keys to, to closing the gap in quality was just using things that, that if you talk to a, a, an electrical engineer off the street, they might know off, off the, like the back of their hand.[00:09:33] Idea 1: Approximation -> Principled Modeling[00:09:33] Dan Fu: But taking some of those properties with how we model dynamical systems in signal processing and then using those ideas to model the inputs, the, the text tokens in, for example a transformer like Next Token Prediction Architecture. So some of those early states-based model papers were looking at this relatively, relatively simple recurrent update model that comes from maybe chapter one of a signal processing class.[00:09:59] Dan Fu: But then using [00:10:00] some principle theory about how you should do that recurrent update in order to really get the most that you can out of your hidden state, out of your out of your sequence. So that, that was one key idea for quality and. When this was eventually realized, you started to see a bunch of benchmarks that were pretty sticky for a few years.[00:10:20] Dan Fu: Things like long range arena, some long sequence evaluation benchmarks, There was stuff in time series, time series analysis. They started to, you started to see the quality tick up in meaningful ways. But the other key thing that What's so influential about these states based models is that they also had a key idea about how you can compute these things efficiently.[00:10:45] Dan Fu: So if you go back to your machine learning 101 class where you learned about RNNs, one thing that you may have learned is that they don't paralyze as well as detention, because if you just run them naively, you have to do this kind of sequential update to process new tokens, [00:11:00] whereas in attention, you can process all the tokens in parallel at one time.[00:11:04] Dan Fu: One of the key insights behind the S4 paper was that these recurrent models, you could take them and you could also formulate them as a convolution. And in particular, with a convolution, you could, instead of using a PyTorch conv1d operation, you can compute that with the FFT. And that would give you n log n compute in the in the sequence length n with an operator that was relatively well optimized for modern hardware.[00:11:28] Dan Fu: So those are really, I'd say, the two key ideas in 2022 that started allowing these breakthroughs to happen in these non transformer architectures. So, these ideas about how to principally model sorry, how to model the recurrent updates of a mo of, of a sequence in a principled way, and also these key ideas in how you can compute it efficiently by turning it into a convolution and then scaling it up with the FFT.[00:11:53] Dan Fu: Along those same lines, so afterwards we started putting out some work on specialized kernels, so just [00:12:00] like we have flash attention for transformers, we also have works like flash fft conf, and if you look at these lines of work oftentimes when, whenever you see a new architecture, you see a new primitive one of the, one of the table stakes now is, do you have an efficient kernel so that you can actually get wall clock speed up?[00:12:14] Idea 3: Selection[00:12:14] Dan Fu: So by 2022, We are starting to have these models that had promising quality primitives, but and, and also promising wall clocks. So you could actually see regimes where they were better than transformers in meaningful ways. That being said, there were, there's still sometimes a quality gap, particularly for language modeling.[00:12:33] Dan Fu: And because languages, It's so core to what we do in sequence modeling these days the, the next, the next key idea that I'm going to talk about is this idea of selection mechanisms. And this is basically an idea of, so you have this recurrent state that you're keeping around that just summarizes everything that, that came before.[00:12:50] Dan Fu: And to get a good sequence model, one of the things that you really need to be able to do is have the model learn what's the best way to pick out pieces from that recurrent [00:13:00] state. So one of the, one of the major ideas here in a line of work called H3, Hungry Hungry Hippos, and also these hyena models were One way you can do this is by just adding some simple element wise gates.[00:13:13] Dan Fu: So versions of these ideas have been around for decades. If you squint at the LSTM paper you, you can probably find, find this gating mechanism. But turns out you can take those old ideas, add them into these new. state space models, and then you can see quality start to pick up. If you've heard of the Mamba model, this also takes the selection to the next level by actually making some changes in that fundamental recurrent state space.[00:13:40] Dan Fu: So, it's not only just this gating that happens around the SSM layer, but also you can actually make The ABCD matrices of your state space model, you can make them data dependent, which will allow you to even better select out different pieces from your hidden state depending on what you're seeing. I'll also point out if you look at the [00:14:00] bottom right of this figure, there's this little triangle with a GPU SRAM, GPU HBM, and this, this is just continuing that trend of when you have a new architecture you, you, you also release it with a kernel to, to, to show that it is hardware efficient, that it, that it can be hardware efficient on modern hardware.[00:14:17] Dan Fu: The, the, one of the next cool things that happened is once we had this understanding of these are the basic pieces, these are the basic principles behind some of the sequence models linear attention actually started to come back. So in earlier this year, there was a model called BASED the, from Simran Arora and, and some other folks, that combined a more principled version of linear attention that basically the, the, the, the two second summary is that it used a Taylor approximation of the softmax attention, combined that with a simple sliding window attention and was starting to able, starting to be able to expand the Pareto frontier of how much data can you recall from your sequence, versus how small is your recurrent state size.[00:14:58] Dan Fu: So those orange dots [00:15:00] are, at the top there, are just showing smaller sequences that can recall more memory.[00:15:07] Just Read Twice[00:15:07] Dan Fu: And the last major idea I think that has been influential in this line of work and is very relatively late breaking just a few months ago, is just the basic idea that when you have these models that are fundamentally more efficient in the sequence length, you maybe don't want to prompt them or use them in exactly the same way.[00:15:26] Dan Fu: So this was a really cool paper called Just Read Twice, also from Simran. That basically said, hey, all these efficient models can process tokens so much more efficiently than transformers that they can sometimes have unfair advantages compared to a simple transformer token. So, or sorry, a simple transformer model.[00:15:44] Dan Fu: So take, for example the standard, the standard use case of you have some long document, you're going to pass it in as input, and then you're going to ask some question about it. One problem you might imagine for a recurrent model where you have a fixed state size is, let's say that [00:16:00] you're. Article is very long, and you're trying to ask about some really niche thing.[00:16:04] Dan Fu: You can imagine it might be hard for the model to know ahead of time what information to put into the hidden state. But these, these, these models are so much more efficient that you can do something really stupid, like, you can just put the document write down the document, write down the question, write down the document again, and then write down the question again, and then this time, the second time that you go over that document, you know exactly what to look for.[00:16:25] Dan Fu: And the cool thing about this is, so this is, And this this results in better quality, especially on these recall intensive tasks. But the other interesting thing is it really takes advantage of the more efficient architectures that, that we're having here. So one of the other, I think, influential ideas in this line of work is if you change the fundamental compute capabilities of your model and the way that it scales, you can actually start to query it at test time differently.[00:16:51] Idea 4: Test Time Compute[00:16:51] Dan Fu: And this actually, of course, goes back to those slides on test time compute. So while everybody's looking at, say, test time compute for big transformer models, [00:17:00] I think potentially a really interesting research question is, how can you take those and how does it change with this new next generation of models?[00:17:09] Dan Fu: So the, I'll just briefly summarize what some of those key ideas were and then talk and then show you briefly kind of what the state of the art is today. So, so the four key ideas are instead of just doing a simple linear attention approximation, instead take ideas that we know from other fields like signal processing, do a more principled approach to your modeling of the sequence.[00:17:32] Idea 2: Hardware & Kernel Support[00:17:32] Dan Fu: Another key idea throughout all these lines of work is you really want. Hardware and kernel support from day one. So, so even if your model is theoretically more efficient if somebody goes and runs it and it's two times slower one of the things that, that we've learned is that if, if you're in that situation, it's, it's just gonna be dead on arrival.[00:17:49] Dan Fu: So you want to be designing your architectures one of the key, key machine learning ideas that has been important for the quality is just making sure that you encode different ways that you can [00:18:00] select from your hidden state and, and really focus on that as a key decider of quality. And finally, I think one of the, the, the emerging new, new things for, for this line of work and something that's quite interesting is, What are the right test time paradigms for these models?[00:18:15] Dan Fu: How do they change relative to relative to what you might do for a standard transformer? I'll briefly end this section. So I've labeled this slide where we are yesterday because Eugene is going to talk about some new models that he released literally this morning. But as of yesterday, some of the really cool results out of the, these efficient alternative models were so AI2 trained this hybrid MOE called Jamba.[00:18:40] Dan Fu: That, that, that seems, that is currently the state of the art for these non transformer architectures. There's this NVIDIA and MIT put out this new diffusion model called SANA recently that one of their key key observations is that you can take a standard diffusion transformer diffusion model, replace the layers with linear [00:19:00] attention, and then that lets you scale to much larger much larger images, much, much Much larger sequences more efficiently.[00:19:07] Dan Fu: And and one thing that I don't think anybody would have called when a few years ago is that one of those gated SSM, gated states based models ended up on the cover of Science because a great group of folks went and trained some DNA models. So that's Michael Polley, Eric Yuen from from Stanford and the Arc Institute.[00:19:26] Dan Fu: So it's, we're really at an exciting time in 2024 where these non transformer, post transformer architectures are showing promise across a wide range. Across a wide range of, of modalities, of applications, and, and of tasks. And with that, I'll pass it on to Eugene, who can tell you a little bit about the latest and greatest with RWKV.[00:19:49] RWKV vs SSMs[00:19:49] Eugene Cheah: So, that's useful? Yeah. You're talking to here. Oh, I'm talking to here. Okay. So, yeah, two streams. Yeah. So, I think one common questions that we tend to get asked, right, is what's the difference between [00:20:00] RWKV and state space? So I think one of the key things to really understand, right the difference between the two groups, right, is that we are actually more like an open source, random internet meets academia kind of situation.[00:20:11] Eugene Cheah: Like, most of us never wrote any paper, but we, we basically look at RNNs and linear intention when intention is all you need came out, and then we decided to like, hey there is a quadratic scaling problem. Why don't we try fixing that instead? So, so, so we end up developing our own branch, but we end up sharing ideas back and forth.[00:20:30] Eugene Cheah: So, and, and we do all this actively in Discord, GitHub, etc. This was so bad for a few years, right, that basically, the average group's H index was so close to zero, right, Illuter. ai actually came in and helped us write our first paper. Great, now our H index is now three, apparently. So, so, so, but, but the thing is, like, a lot of these experiments led to results, and, and, essentially, essentially, we we took the same ideas from linear attention, [00:21:00] and we built on it.[00:21:01] Eugene Cheah: So, to take a step back into, like, how does RWKB handle its own attention mechanic and achieve the same goals of, like, O and compute, respectively, and in focus of our overall goal to make AI accessible to everyone, regardless of language, nation, or compute, that's our goal. We actually train our models primarily on over a hundred languages, which is another topic altogether.[00:21:23] Eugene Cheah: And our goal is to train to even 200 languages to cover all languages in the world. But at the same time, we work on this architecture, To lower the compute cost so that people can run it on Raspberry Pis and on anything. So, how did RWKB break the dependency of LSTM token flow? Because I think to understand architecture, right, it's probably easier to understand it from the RNN lens.[00:21:46] Eugene Cheah: Because that's where we built on. We all, we all state space kind of like try to, try to start anew and took lessons from that and say, So there's a little bit of divergence there. And AKA, this our version of linear attention. So to take step back [00:22:00] all foundation models, be it transformers or non transformers at a very high level, right?[00:22:05] Eugene Cheah: Pumps in the token. I mean, text that things into embeddings and go through a lot of layers. Generate a lot of states where the QKV cache or be iron in states or RW KB states. And outputs and embedding, they are not the same thing. And we just take more layers and more embeddings. And somehow that magically works.[00:22:23] Eugene Cheah: So, if you, if you remember your ancient RNN lessons which we, which we, which we we call best learning these days the general idea is that you have the embedding information flowing all the way up, and when, and you take that information and you flow it back down, and then you process it as part of your LSTM layers.[00:22:41] Eugene Cheah: So, this is how it generally works. Kapati is quoted saying that RNNs are actually unreasonably effective. The problem is this is not scalable. To start doing work on the second token, you need to wait for the first token. And then you need to, and likewise for the third token and fourth token, yada yada.[00:22:55] Eugene Cheah: That is CPU land, not GPU land. So, so, so, you [00:23:00] can have a H100 and you can't even use 1 percent of it. So, so that's kind of why RNNs didn't really take off in the direction that we wanted, like, billions of parameters when it comes to training. So, what did RDAP KV version 0 do? Boom. We just did the dumbest, lamest thing.[00:23:13] Eugene Cheah: Sorry, this is the bottleneck for RNN. We did the dumb thing of removing that line. And it kind of worked. It trained. It sucked, but it kind of worked. Then we were like, hey, then no one cared because the loss was crap, but how do we improve that? And that's essentially where we move forward, because if you see this kind of flow, right, you can actually get your GPU saturated quickly, where it essentially cascades respectively.[00:23:41] Eugene Cheah: So I'm just waiting for this to loop again. So it's like, once you get your first layer, your token to be computed finish. You start to cascade your compute all the way until you are, Hey, I'm using 100 percent of the GPU. So we, we worked on it, and we started going along the principle of that as long as we keep this general architecture [00:24:00] where, where we can cascade and, and be highly efficient with our architecture, nothing is sacred in our architecture.[00:24:06] Eugene Cheah: And we have done some crazy ideas. In fact, you ask us, if you ask me to explain some things in the paper, right, officially in the paper, I'll say we had this idea and we wrote it this way. The reality is someone came with a code, we tested it, it worked, and then we rationalized later. So, so the general[00:24:24] RWKV Arch[00:24:24] Eugene Cheah: The idea behind rwkbr is that we generally have two major blocks that we do.[00:24:30] Eugene Cheah: We call time mix and channel mix. And time mix generally handles handles long term memory states, where essentially, where essentially where we apply the matrix multiplication and Cilu activation functions into processing an input embedding and an output embedding. I'm oversimplifying it because this, This calculation changed every version and we have, like, version 7 right now.[00:24:50] Eugene Cheah: ChannelMix is similar to Base in the sense that it does shorter term attention, where it just looks at the sister token, or the token before it, because [00:25:00] there's a shift in the token shift matrix. I don't really want to go too much into the papers itself, because, like, we do have three papers on this.[00:25:09] Eugene Cheah: Basically, RWKB, RNN for the transformer, ERA, Ego and Pinch, RWKB, Matrix Value State. This is the updated version 5, version 6. And Goldfinch is our, is, is, is, is our hybrid model respectively. We are writing the paper already for V seven and which is, which is for R wk V seven. Called, named Goose, or architectures are named by Bird.[00:25:30] Eugene Cheah: And, I'm going to cover as well, qrwkb, and mama100k, and rwkb, and Where did that lead to? Great! Because we are all GPU poor and to be clear, like, most of this research is done, like, only on a handful H100s, which I had one Google researcher told me that was, like, his experiment budget for a single researcher.[00:25:48] Eugene Cheah: So, our entire organization has less compute than a single researcher in Google. So We, we, one of the things that we explored into was to how do we convert transformer models instead? Because [00:26:00] someone already paid that billion dollars, a million dollars onto training, so why don't we take advantage of those weights?[00:26:05] Eugene Cheah: And, and to, I believe, together AI worked on the lockets for, for the Lambda side of things, and, and we took some ideas from there as well, and we essentially did that for RWKB.[00:26:15] QWRKWv6 launch[00:26:15] Eugene Cheah: And that led to, Q RWKB6, which we just dropped today, a 32 bit instruct preview model, where we took the Quen 32 bit instruct model, freeze the feedforward layer, remove the QKB attention layer, and replace it with RWKB linear layers.[00:26:32] Eugene Cheah: So to be clear, this means we do not have the rwkv channel mix layer, we only have the time mix layer. But but once we do that, we train the rwkv layer. Important is that the feedforward layer needs to be frozen, so the new attention can be learned. And then we unfreeze the feedforward layer, and train all the layers together with a custom learning rate schedule, so that they can learn how to work together.[00:26:54] Eugene Cheah: The end result, surprisingly, And, to be honest, to the frustration of the R. W. [00:27:00] KV MOE team, which ended up releasing the model on the same day, was that, with just a few hours of training on two nodes, we managed to get it to be on par, kind of, with the original QUAN32B model. So, in fact, when the first run, right, that completely confused us, it was like, and I was telling Daniel Goldstein, Smirky, who kind of leads most of our research coordination, When you pitched me this idea, you told me at best you'll get the same level of performance.[00:27:26] Eugene Cheah: You didn't tell me the challenge and score and Winograd score will shoot up. I don't know what's happening there. But it did. MMLU score dropping, that was expected. Because if you think about it, when we were training all the layers, right, we were essentially Like, Frankenstein this thing, and we did brain damage to the feedforward network layer 2 with the new RWKB layers.[00:27:47] Eugene Cheah: But, 76%, hey, somehow it's retained, and we can probably further train this. We didn't even spend more than 3 days training this, so there's a lot more that can be done, hence the preview. This brings up [00:28:00] a big question, because We are already now in the process of converting to 7TB. We are now, this is actually extremely compute efficient to test our attention mechanic.[00:28:10] Eugene Cheah: It's like, it becomes a shortcut. We can, we are already planning to do our version 7 and our hybrid architecture for it. Because we don't need to train from scratch. And we get a really good model out of it. And the other thing that is uncomfortable to say is that because we are doing right now on the 70b is that if this scales correctly to 128k context length, I'm not even talking about a million 128, majority of enterprise workload today is just on 70b at under 32k context length.[00:28:41] Eugene Cheah: That means if this works and the benchmark matches it, It means we can replace the vast majority of current AI workload, unless you want super long context. And then sorry, can someone give us more GPUs? Because we do need the VRAM for super long context, sadly. So yeah, that's what we are working on, and essentially, [00:29:00] we are excited about this to just push it further.[00:29:02] Eugene Cheah: And this conversion process, to be clear, I don't think it's going to be exclusive to RWKB. It probably will work for Mamba as well, I don't see why not. And we will probably see more ideas, or more experiments, or more hybrids, or Yeah, like, one of the weirdest things that I wanted to say outright, and I confirmed this with the Black Mamba team and the Jamba team, which because we did the GoFinch hybrid model, is that none of us understand why a hard hybrid with a state based model to be R.[00:29:28] Eugene Cheah: QA state space and transformer performs better when, than the baseline of both. It's like, it's like when you train one, you expect, and then you replace, you expect the same results. That's our pitch. That's our claim. But somehow when we jam both together, it outperforms both. And that's like one area of emulation that, like, we only have four experiments, plus four teams, that a lot more needs to be done.[00:29:51] Eugene Cheah: But, but these are things that excite me, essentially, because that is what it's potentially we can move ahead for. Which brings us to what comes next.[00:30:00] What's next[00:30:00] [00:30:00][00:30:00] Dan Fu: So, this part is kind of just some, where we'll talk a little bit about stuff that, that we're excited about. Maybe have some wild speculation on, on what, what's, what's coming next.[00:30:12] Dan Fu: And, of course this is also the part that will be more open to questions. So, a couple things that, that I'm excited about is continued hardware model co design for, for these models. So one of the things that we've put out recently is this library called ThunderKittens. It's a CUDA library.[00:30:29] Dan Fu: And one of the things that, that we found frustrating is every time that we built one of these new architectures, and I'm sure you had the exact same experience, we'd have to go and spend two months in CUDA land, like writing these, these new efficient things. And. If we decided to change one thing in PyTorch, like one line of PyTorch code is like a week of CUDA code at least.[00:30:47] Dan Fu: So one of our goals with, with a library like Thunderkitten, so we, we just broke down what are the key principles, what are the key hardware things what are the key, Compute pieces that you get from the hardware. So for example on [00:31:00] H100 everything is really revolves around a warp group matrix multiply operation.[00:31:06] Dan Fu: So you really want your operation to be able to split into relatively small matrix, matrix multiply operations. So like multiplying two 64 by 64 matrices, for example. And so if you know that ahead of time when you're designing your model, that probably gives you you know, some information about how you set the state sizes, how you set the update, how you set the update function.[00:31:27] Dan Fu: So with Thunderkittens we basically built a whole library just around this basic idea that all your basic compute primitives should not be a float, but it should be a matrix, and everything should just be matrix compute. And we've been using that to, to try to both re implement some existing architectures, and also start to design code.[00:31:44] Dan Fu: Some new ones that are really designed with this core with a tensor core primitive in mind. Another thing that that we're, that at least I'm excited about is we, over the last four or five years, we've really been looking at language models as the next thing. But if you've been paying [00:32:00] attention to Twitter there's been a bunch of new next generation models that are coming out.[00:32:04] Dan Fu: So there, there are. So, video generation models that can run real time, that are supported by your mouse and your keyboard, that I'm told if you play with them that, you know, that they only have a few seconds of memory. Can we take that model, can we give it a very long context length so that you could actually maybe generate an entire game state at a time?[00:32:25] Dan Fu: What does that look like for the model? You're certainly not going to do a giant quadratic attention computation to try to run that. Maybe, maybe use some of these new models, or some of these new video generation models that came out. So Sora came out I don't know, two days ago now. But with super long queue times and super long generation times.[00:32:43] Dan Fu: So that's probably a quadratic attention operation at the, at the bottom of it. What if we could remove that and get the same quality, but a lot faster generation time? Or some of the demos that we saw from Paige earlier today. You know, if I have a super long conversation with my [00:33:00] Gemini bot, what if I wanted to remember everything that it's seen in the last week?[00:33:06] Dan Fu: I mean, maybe you don't for personal reasons, but what if I did, you know? What does that mean for the architecture? And I think, you know, that's certainly something I'm pretty excited about. I'm sure you're excited about it too. So, I think we were supposed to have some hot takes, but I honestly don't remember what our hot takes were.[00:33:21] Hot Takes - does anyone really need long context?[00:33:21] Eugene Cheah: Yeah, including the next slide. Hot takes, yes, these are our[00:33:25] Dan Fu: hot takes.[00:33:25] Eugene Cheah: I think the big one on Twitter that we saw, that we shared, was the question is like, is RAG relevant? In the case of, like, the future of, like, state based models?[00:33:38] Dan Fu: Let's see, I haven't played too much with RAG. But when I have. I'll say I found it was a little bit challenging to do research on it because we had this experience over and over again, where you could have any, an embedding model of any quality, so you could have a really, really bad embedding model, or you could have a really, really [00:34:00] good one, By any measure of good.[00:34:03] Dan Fu: And for the final RAG application, it kind of didn't matter. That's what I'll say about RAG while I'm being recorded. I know it doesn't actually answer the question, but[00:34:13] Eugene Cheah: Yeah, so I think a lot of folks are like, extremely excited of the idea of RWKB or State Space potentially having infinite context.[00:34:21] Eugene Cheah: But I think the reality is that when we say infinite context, we just mean a different kind of infinite context, or you, or as it's previously covered, you need to test the model differently. So, think of it more along the lines of the human. Like, I don't remember what I ate for breakfast yesterday.[00:34:37] Eugene Cheah: Yeah, that's the statement that I'll say. And And we humans are not quadratic transformers. If we did, if let's say we increased our brain size for every second we live, we would have exploded by the time we are 5 years old or something like that. And, and I think, I think basically fundamentally for us, right, be it whether we, regardless of whether RWKB, statespace, XLSTM, [00:35:00] etc, our general idea is that instead of that expanding state, that increase in computational cost, what if we have a fixed state size?[00:35:08] Eugene Cheah: And Information theory detects that that fixed state size will have a limit. Just how big of a limit is a question, like, we, like, RWKB is running at 40 megabytes for, for its state. Its future version might run into 400 megabytes. That is like millions of tokens in, if you're talking about mathematically, the maximum possibility.[00:35:29] Eugene Cheah: It's just that I guess we were all more inefficient about it, so maybe we hit 100, 000. And that's kind of like the work we are doing, trying to like push it and maximize it. And that's where the models will start differing, because it will choose to forget things, it will choose to remember things. And that's why I think that there might be some element of right, but it may not be the same right.[00:35:49] Eugene Cheah: It may be the model learn things, and it's like, hmm, I can't remember that, that article. Let me do a database search, to search. Just like us humans, when we can't remember the article in the company. We do a search on Notion. [00:36:00][00:36:00] Dan Fu: I think something that would be really interesting is if you could have facts that are, so right now, the one intuition about language models is that all those parameters are around just to store random facts about the world.[00:36:14] Dan Fu: And this intuition comes from the observation that if you take a really small language model, it can do things like talk to you, or kind of has like the The style of conversation, it can learn that, but where it will usually fall over compared to a much larger one is it'll just be a lot less factual about things that it knows or that it can do.[00:36:32] Dan Fu: But that points to all those weights that we're spending, all that SGD that we're spending to train these models are just being used to store facts. And we have things like databases that are pretty good at storing facts. So I think one thing that would be really interesting is if we could actually have some sort of outside data store that a language model can can look at that that maybe is you know, has has some sort of gradient descent in it, but but would be quite interesting.[00:36:58] Dan Fu: And then maybe you could edit it, delete [00:37:00] facts, you know, change who's president so that it doesn't, it doesn't get lost.[00:37:04] Vibhu: Can we open up Q& A and hot takes for the audience? I have a hot take Q& A. Do these scale? When, when 405B state space model, RAG exists, no one does long context, who's throwing in 2 million token questions, hot takes?[00:37:24] Dan Fu: The, the who's throwing in 2 million token question, I think, is, is a really good question. So I actually, I was going to offer that as a hot take. I mean, my hot take was going to be that long context doesn't matter. I know I just gave a whole talk about it, but you know, what, what's the point of doing research if you can't, you know, play both sides.[00:37:40] Dan Fu: But I think one of the, so I think for both of us, the reason that we first got into this was just from the first principled questions of there's this quadratic thing. Clearly intelligence doesn't need to be quadratic. What is going on? Can we understand it better? You know, since then it's kind of turned into a race, which has [00:38:00] been exciting to watch, like, how much context you can take in.[00:38:03] Dan Fu: But I think it's right. Nobody is actually putting in a two million context prompt into these models. And, and, you know, if they are, maybe we can go, go You know, design a better model to do that particular thing. Yeah, what do you think about that? So you've also been working on this. Do you think long context matters?[00:38:19] Eugene Cheah: So I'm going to burn a bit. How many of you remember the news of Google Gemini supporting 3 million contacts, right? Raise your hand.[00:38:28] Vibhu: Yeah, 2 million.[00:38:29] Eugene Cheah: Oh, it's 2 million.[00:38:31] Eugene Cheah: Yeah, how many of you actually tried that? See?[00:38:34] Vibhu: I use it a lot. You? You work for MindsTV. I use it a lot.[00:38:41] Eugene Cheah: So, for some people that has used, and I think, I think that's the, that's might be, like, this is where my opinion starts to differ, because I think the big labs may have a bigger role in this, because Like, even for RWKB, even when we train non contacts, the reason why I say VRAM is a problem is that because when we did the, we need to backprop [00:39:00] against the states, we actually need to maintain the state in between the tokens by the token length.[00:39:05] Eugene Cheah: So that means we need to actually roll out the whole 1 million contacts if we are actually training 1 million. Which is the same for transformers, actually, but it just means we don't magically reuse the VRAM consumption in the training time space. So that is one of the VRAM bottlenecks, and I'm neither OpenAI nor Google, so donate GPUs if you have too much of them.[00:39:27] Eugene Cheah: But then, putting it back to another paradigm, right, is that I think O1 style reasoning might be actually pushing that direction downwards. In my opinion, this is my partial hot take is that if, let's say you have a super big model, And let's say you have a 70B model that may take double the tokens, but gets the same result.[00:39:51] Eugene Cheah: Strictly speaking, a 70B, and this is even for transformer or non transformer, right? We we'll take less less resources than that 400 B [00:40:00] model, even if it did double the amount thinking. And if that's the case, and we are still all trying to figure this out, maybe the direction for us is really getting the sub 200 B to be as fast as efficient as possible.[00:40:11] Eugene Cheah: We a very efficient architecture that some folks happen to be working on to, to just reason it out over larger and larger context thing.[00:40:20] Question: Yeah. One thing I'm super interested in is. Models that can watch forever? Obviously you cannot train something on infinite context length. How are y'all thinking about that, where you run on a much longer context length than is possible to train on?[00:40:38] Dan Fu: Yeah, it's a, it's a great question. So I think when I think you guys probably had tweets along these lines, too. When we first started doing these things, because these are all recurrent models in theory you could just run it forever. You could just run it forever. And at the very least it won't, it won't like error out on your crash.[00:40:57] Dan Fu: There's another question of whether it can actually [00:41:00] use what it's seen in that infinite context. And I think there, so one place where probably the research and architectures ran faster Then another research is actually the benchmarks for long context. So you turn it on forever. You want to do everything or watch everything.[00:41:16] Dan Fu: What is it that you actually wanted to do? Can we actually build some benchmarks for that? Then measure what's happening. And then ask the question, can the models do it? Is there something else that they need? Yeah, I think that if I were to turn back the clock to 2022, that's probably one of the things I would have done differently, which would have been actually get some long context benchmarks out at the same time as we started pushing context length on all these models.[00:41:41] Eugene Cheah: I will also say the use case. So like, I think we both agree that there's no Infinite memory and the model needs to be able to learn and decide. I think what we have observed for, I think this also fits the state space model, is that one of the key advantages of this alternate attention mechanic that is not based on token position is that the model don't suddenly become crazy when you go past the [00:42:00] 8k training context tank, or a million context tank.[00:42:03] Eugene Cheah: It's actually still stable. It's still able to run, it's still able to rationalize. It just starts forgetting things. But some of these things are still there in latent memory. Some of these things are still somewhat there. That's the whole point of why reading twice works. Things like that. And one of the biggest pushes in this direction is that I think both Statespace and RWKB have Separate papers by other researchers where they use this architecture for time series data.[00:42:26] Eugene Cheah: Weather modeling. So, you are not asking what was the weather five days ago. You're asking what's the weather tomorrow based on the infinite length that we, as long as this Earth and the computer will keep running. So, so, and they found that it is like, better than existing, like, transformer or existing architecture in modeling this weather data.[00:42:47] Eugene Cheah: Control for the param size and stuff. I'm quite sure there are people with larger models. So, so there are things that, that in this case, right, there is future applications if your question is just what's next and not what's 10 years ago.[00:42:59] Dan Fu: Thanks so [00:43:00] much for having us. Get full access to Latent Space at www.latent.space/subscribe

eisbaerlin.de der Podcast - mit Jule Daniel, Hannes und KevinIn Episode 3 der Saison 2024/2025 des eisbaerlin.de Podcasts müssen wir leider krankheitsbedingt auf Jule verzichten. Dafür springt Ex Eisbären-Pressesprecher und Bend Your Knees Host Daniel Goldstein ein. Mit Daniel sprechen Kevin und Hannes u.a. über:Thomas Greiss in Frankfurtdie Fanproteste zu den Donnerstagsspielenden anstehenden DFEL Startdie bisherigen Spiele der Eisbären inkl. der Bannerzeremonie beim ersten Heimspielden PinktoberWeitere Infos von eisbaerlin.de findet ihr u.a. hier:Website: https://eisbaerlin.de/blog/Instagram: https://www.instagram.com/eisbaerlin/Facebook: https://www.facebook.com/eisbaerlin.fanclubTwitter: https://twitter.com/eisbaerlinBlue Sky: https://bsky.app/profile/eisbaerlinwbn.bsky.socialThreads: https://www.threads.net/@eisbaerlinFolgt Jule auf Twitter: https://twitter.com/Julschke97Folgt Hannes auf Twitter: https://twitter.com/h_modesFolgt Kevin auf Twitter: https://twitter.com/eisbaerlin_WBNFolgt Daniel auf Twitter: https://x.com/dfg193 Hosted on Acast. See acast.com/privacy for more information.

This month, JHLT: The Podcast reissues our September 2023 tribute to former Editor-in-Chief, Dr. Daniel R. Goldstein. Dr. Goldstein stepped down from his role for health reasons in July 2023; he had been diagnosed with an advanced salivary gland malignancy and felt he would be unable to continue serving JHLT and the International Society for Heart and Lung Transplantation (ISHLT) to his characteristically demanding standard. Sadly, Dr. Goldstein died on 21 May, 2024, at the age of 56, leaving behind his wife, 2 children, an extended family, and a larger universe of colleagues, collaborators, and mentees who greatly benefited from his equanimity, wisdom, and commitment to his passions. The tribute, recorded in August 2023, features Michelle Kittleson, MD, PhD, then-Interim Editor-in-Chief of JHLT; Andrew Gelman, PhD, Deputy Editor at JHLT; Andrew Fisher, FRCP, PhD, past president of ISHLT and Past Chair of the Publications Oversight Committee; Daniel Tyrrell, PhD, a former post-doc of Dr. Goldstein's; and Judy Chen, PhD, a former immunology PhD student in Dr. Goldstein's lab. Two funds were created to allow friends and colleagues to memorialize Dr. Goldstein: the Michigan Biology of Cardiovascular Aging Leadership Development Fund at the Frankel Cardiovascular Center (https://giving.umich.edu/give/393178) and the Adenoid Cystic Carcinoma Foundation (https://accrf.org). The JHLT has also re-published Dr. Goldstein's farewell message in the September 2024 issue of the Journal. You can read it here: https://www.jhltonline.org/article/S1053-2498(24)01741-8/fulltext Follow along at www.jhltonline.org/current, or, if you're an ISHLT member, log in at ishlt.org/journal-of-heart-lung-transplantation.  Don't already get the Journal and want to read along? Join the International Society of Heart and Lung Transplantation at www.ishlt.org for a free subscription, or subscribe today at www.jhltonline.org.    

Bald geht sie endlich wieder los, die DEL! Und wie ihr es gewohnt seid, gibt es auch dieses Jahr eine Vorschau auf alle Teams. Mit unseren Gästen Tina Fröhlich (Südkurier) und Daniel Goldstein (u.a. Tagesspiegel) fliegen wir durch die Liga, schauen auf alle Teams und haben sogar noch ein DFEL-Extra für euch im Gepäck. Eiszeit FM sind Philipp Koehl und Sven Metzger. Du möchtest deinen Podcast auch kostenlos hosten und damit Geld verdienen? Dann schaue auf www.kostenlos-hosten.de und informiere dich. Dort erhältst du alle Informationen zu unseren kostenlosen Podcast-Hosting-Angeboten. kostenlos-hosten.de ist ein Produkt der Podcastbude.Gern unterstützen wir dich bei deiner Podcast-Produktion.

Bald geht sie endlich wieder los, die DEL! Und wie ihr es gewohnt seid, gibt es auch dieses Jahr eine Vorschau auf alle Teams. Mit unseren Gästen Tina Fröhlich (Südkurier) und Daniel Goldstein (u.a. Tagesspiegel) fliegen wir durch die Liga, schauen auf alle Teams und haben sogar noch ein DFEL-Extra für euch im Gepäck. Eiszeit FM sind Philipp Koehl und Sven Metzger. Du möchtest deinen Podcast auch kostenlos hosten und damit Geld verdienen? Dann schaue auf www.kostenlos-hosten.de und informiere dich. Dort erhältst du alle Informationen zu unseren kostenlosen Podcast-Hosting-Angeboten. kostenlos-hosten.de ist ein Produkt der Podcastbude.Gern unterstützen wir dich bei deiner Podcast-Produktion.

Bald geht sie endlich wieder los, die DEL! Und wie ihr es gewohnt seid, gibt es auch dieses Jahr eine Vorschau auf alle Teams. Mit unseren Gästen Tina Fröhlich (Südkurier) und Daniel Goldstein (u.a. Tagesspiegel) fliegen wir durch die Liga, schauen auf alle Teams und haben sogar noch ein DFEL-Extra für euch im Gepäck. Eiszeit FM sind Philipp Koehl und Sven Metzger.

Rural communities looking to improve economic opportunities and retain jobs and community wealth know employee ownership can help them achieve these objectives. In fact, some of the states, towns, and cities with the densest number of employee owned-companies are in primarily rural areas. These companies range from small worker cooperatives in agriculture and energy to medium and large companies in manufacturing and construction, among other sectors. During this panel conversation, we'll hear from elected officials and employee-owned companies in rural areas about the important role of employee ownership in their communities and the lessons they have learned about bringing shared prosperity to rural areas. Speakers include: Daniel Goldstein, Chief Executive Officer, Go ESOP LLC; Executive Fellow, Rutgers Institute for the Study of Employee Ownership and Profit Sharing Molly Hemstreet, Founder and Co-Executive Director, The Industrial Commons; Executive Fellow, Rutgers Institute for the Study of Employee Ownership and Profit Sharing Jenny Levy, Executive Vice President, People, Community & Environment, Hypertherm Michael Williams, CFO, Black & Veatch Jeff Guo, Co-host and Reporter for Planet Money, NPR (moderator) This video comes from the second Employee Ownership Ideas Forum, hosted by the Aspen Institute Economic Opportunities Program and the Rutgers Institute for the Study of Employee Ownership and Profit Sharing. Our 2024 Forum, “Employee Ownership on the Ground,” brought innovative employee share ownership initiatives and speakers from around the country to Washington DC to highlight how this bipartisan approach to improving jobs, wealth creation, and business performance is helping create more equitable economies in states, cities, and rural communities. For clips and highlights from the Forum, subscribe to the Economic Opportunities Program on YouTube: https://www.youtube.com/@aspeneop/ And tune in to our podcast to listen to full discussions on the go: https://podcasters.spotify.com/pod/show/aspeneop For more from the Forum — including videos, photos, audio, transcripts, and additional resources — visit: https://www.aspeninstitute.org/events/employee-ownership-ideas-forum-2024/

Bret Keisling is joined by Daniel Goldstein, one of the leading voices in ESOPs and employee ownership today, who shares his insights on the transformative impact of ESOPs, fostering employee ownership, enhancing community welfare, and promoting business sustainability. Daniel addresses challenges faced by ESOPs, the significance of education and communication within organizations, and the role of employee ownership in addressing racial wealth disparities. Daniel emphasizes the importance of implementing EO across diverse industries and communities and its importance in rural areas for driving long-term employee wealth, business resilience, and economic stability. For more information, visit our website for this episode at https://www.theesoppodcast.com/post/278-daniel-goldstein-from-esop-ceo-to-advocate

Daniel Goldstein describes his work on this issue.

Bret Keisling is joined by Daniel Goldstein, retired president and CEO of Folience, Inc. [https://www.folience.com/], an ESOP holding company. Daniel was on a panel at this week's Employee Ownership Ideas Forum in a session called "Hometown Ownership: The Impact of Employee Ownership in Rural Communities."  In this excerpt from an upcoming EsOp Podcast episode, Daniel discusses how ESOPs in rural areas provide economic opportunities, support multi-generational employment, and use innovative technology to upskill employees.  The Employee Ownership Ideas Forum was presented by the Aspen Institute and the Rutgers Institute for the Study of Employee Ownership and Profit Sharing on April 9-10, 2024, in Washington D.C. Head to our website for the transcript and video for this episode, with links to all the resources mentioned: https://www.theesoppodcast.com/post/mini-cast-261-daniel-goldstein-on-eo-and-rural-america

Welcome to The Private Equity Podcast! This episode dives into the transformative world of employee ownership, focusing on the Employee Stock Ownership Plan and how it can drive employee engagement and productivity. Guest Daniel Goldstein, former CEO of Folience and an advocate for employee ownership, shares insights into integrating ESOP within a private equity model, fostering workplace culture, and exploring partial ownership benefits.Breakdown:[00:00] Introduction: Welcome to the podcast and introduction of Daniel Goldstein.[00:21] Exploring Employee Ownership: Discussion on the benefits of workplace culture improvement and the contrast between U.S. ESOPs and UK EOTs.[01:13] Daniel Goldstein's Background: Overview of Goldstein's career and his approach to employee ownership within the Private Equity model.[02:08] Employee Ownership as a Continuum: Insights into the relationship between family-owned businesses, employee-owned businesses, and Private Equity.[03:33] Advantages of Employee Ownership: Benefits of aligning employee interests with company success.[04:32] Folience's Unique Model: How Folience operates with a fully employee-owned Private Equity model.[05:55] Growth and Diversification Strategy: The strategy behind acquiring businesses and integrating them into the Folience portfolio.[07:50] Tax Efficiency and Ownership Structures: The tax benefits and strategic advantages of employee ownership models.[08:49] Decision Making in an ESOP: Governance and decision-making within an ESOP structure.[11:10] Shared Ownership and Transparency: The role of shared ownership in fostering workplace transparency and collaboration.[13:29] Employee Engagement Without Full Ownership: Strategies to improve engagement without full employee ownership.[14:27] Liquidity Events and Portfolio Company Sales: Effects on employee owners during liquidity events and sales.[18:18] Implementing Partial Employee Ownership: The potential benefits of partial employee ownership.[24:28] The Power of Incentivization: How financial incentives can drive employee productivity.[32:29] The Silver Tsunami: Challenges and opportunities in the generational transfer of wealth and business ownership.[37:16] Contacting Daniel Goldstein: Information for reaching out to Daniel for more discussion on employee ownership.[38:59] Episode Conclusion: Summary and thanks to Daniel Goldstein, with a reminder on how to reach Raw Selection for hiring support.To connect with Daniel Goldstein, you can visit his LinkedIn profile.Thank you for tuning in!To get the newest Private Equity episodes, you can subscribe on iTunes or Spotify here.Lastly, if you have any feedback on the podcast or want to reach out to Alex with any questions, send an email to alex.rawlings@raw-selection.com.

Join us in this week's episode to hear all about renewable energies and their impact on our future from intern Daniel Goldstein. Daniel is a senior electrical engineering major at the University of Nebraska-Lincoln and a Pathway intern on the Solar track. Stick around for the latter half of the show to hear all about cooking. In this week's segment, hosts Rachel and Brayton discuss cooking tips, recipes, and where to find cooking inspiration. If you have a favorite recipe, we'd love to hear it in the comments below!

Join us in this week's episode to learn more about the different types of renewable energy and how they are impacting our future. In this discussion, we interview Solar Pathway intern and senior electrical engineering major Daniel Goldstein to learn more about these industries. Todays segment is a treat, tune in to hear from hosts Rachel and Brayton all about cooking!

Mein langjähriger Küchenmaschinenhändler Daniel Goldstein hat seit geraumer Zeit auf Anregung meiner Blogleser auch die Knetmaschinen Wilfa Probaker und Graef Myestro im Programm. Im Gespräch mit mir geht er auf die Vor- und Nachteile beider Maschinen ein und erklärt, welches Gerät für welchen Anwendungsfall geeignet ist und am besten in welche Küche passt - auch im Vergleich zu WMF, Kenwood und KitchenAid.

龍年HIGH起來！新光三越《龍舞卡利HIGH》獨享7%回饋

Bei Eiszeit FM pflegen wir Traditionen und deshalb darf die obligatorische große Saison-Vorschau natürlich nicht fehlen. Zusammen mit Martin Wimösterer (u.a. Eishockey News) und Daniel Goldstein (u.a. Bend your knees) besprechen wir die Stärken und Schwächen der 14 DEL-Teams. Wenn ihr darüber hinaus wissen wollt, was "Ich liebe dich, Karawanken-Express" auf slowenisch heißt, warum kaum ein Weg in der kommenden Saison an den Kölner Haien vorbeiführt und welcher Adler-Zugang der wichtigste ist, dann seid ihr hier genau richtig. Eiszeit FM sind Philipp Koehl und Sven Metzger. Du möchtest deinen Podcast auch kostenlos hosten und damit Geld verdienen? Dann schaue auf www.kostenlos-hosten.de und informiere dich. Dort erhältst du alle Informationen zu unseren kostenlosen Podcast-Hosting-Angeboten. kostenlos-hosten.de ist ein Produkt der Podcastbude.Gern unterstützen wir dich bei deiner Podcast-Produktion.

Bei Eiszeit FM pflegen wir Traditionen und deshalb darf die obligatorische große Saison-Vorschau natürlich nicht fehlen. Zusammen mit Martin Wimösterer (u.a. Eishockey News) und Daniel Goldstein (u.a. Bend your knees) besprechen wir die Stärken und Schwächen der 14 DEL-Teams. Wenn ihr darüber hinaus wissen wollt, was "Ich liebe dich, Karawanken-Express" auf slowenisch heißt, warum kaum ein Weg in der kommenden Saison an den Kölner Haien vorbeiführt und welcher Adler-Zugang der wichtigste ist, dann seid ihr hier genau richtig. Eiszeit FM sind Philipp Koehl und Sven Metzger.

This podcast is a commentary and does not contain any copyrighted material of the reference source. We strongly recommend accessing/buying the reference source at the same time. ■Reference Source https://www.ted.com/talks/daniel_goldstein_the_battle_between_your_present_and_future_self ■Post on this topic (You can get FREE learning materials!) https://englist.me/148-academic-words-reference-from-daniel-goldstein-the-battle-between-your-present-and-future-self-ted-talk/ ■Youtube Video https://youtu.be/VrvDtkuluRc (All Words) https://youtu.be/aZG-B2F6Bj4 (Advanced Words) https://youtu.be/KgLqEROhekc (Quick Look) ■Top Page for Further Materials https://englist.me/ ■SNS (Please follow!)

An dieser Stelle dokumentieren wir den Talk zum Thementag Fraueneishockey mit Eisbärinnen-Mannschaftsleiter Torsten Szyska und Co-Trainerin Kathrin Fring moderiert von Daniel Goldstein und Peter Breitenberger. Der Tag war eine Veranstaltung vom EBB-Fanclub Block 417 & Freunde.

Daniel Goldstein discusses cost effectiveness of cancer therapy. 

Daniel Goldstein war über viele Jahre Pressesprecher der Eisbären Berlin und ein bekanntes Gesicht der DEL. Vor zwei Jahren trennten sich die Wege Daniels und der Eisbären. Nach einem kurzen Abstecher zum DEB ging Daniel in die politische Kommunikation. Er arbeitet heute für die Fraktion der Partei "Die Linke" im Brandenburger Landtag. Mit Christoph Ullrich redet er darüber, was Politik und Eishockey gemeinsam haben. Und auch über Schlägereien im Eishockey - wie jüngst bei Nürnberg gegen Frankfurt. Viel Spaß beim Hören.

“Your future depends on what you do today.” - Mahatma Ghandi Future Self is the person who you will be years from now. It is the person you intend to be. Do you know who your future self is, or better yet, have you decided who you are going to be in the future without thinking of any limitations? Is he or she a healthier, wealthier, happier, better person than you are today, or would he be the same as the person you are today, only physically older? Why is it important to know the person you are becoming and, more significantly, is it possible for you to have a relationship with your future self? If so, what good would it do? Tune in and let me inspire and show you how to be the person you have been eyeing to be, and realize the endless possibilities that you can have for yourself. Have a relationship with your future self and be the best person you can be, be the change the world wants you to be. Highlights: ⚡️ We are driven by pain and our view of the future and what we do today impacts tomorrow. ⚡️ When people identify with the person they want to become, they become that person. ⚡️ We must believe that our future self already exists and have a relationship with him and be guided by him. ⚡️ Failing to be the person you want to be results in depression, anxiety, and frustration. ⚡️ Reasons why most people are living a future they do not want: pursuing the ideals of somebody else and not who they want to become choosing exactly who they are today because that is who they think they are failing to prospect, losing the ability to be a free thinker, and not realizing the choices they have do not have a relationship with and empathy for their future self not getting in the game they do not see the benefit of doing something despite the challenges Important stories:

Während ein Teil der meinungsstarken Menschen die olympischen Spiele und deren Vergabe nach China mit guten Gründen komplett verteufelt und andere meinen, man sollte sich nur auf den Sport konzentrieren, fragte sich Daniel Goldstein vor allem, wie ist es eigentlich jenseits dessen. Wie lebt es sich in China? Was ist Peking für eine Stadt? Und, was passiert dort eigentlich abseits der großen Scheinwerfer? Und all diese Fragen kann Torsten Weidemann, Eishockeyfan, der vier Jahre in China lebte beantworten. Ein Gespräch über China, Eishockey und mehr.

Wenn alle im Olympiafieber sind, dann federn wir ordentlich durch und lassen uns natürlich auch anstecken. Kathrin Fring, André Rankel und Daniel Goldstein sprechen über die olympischen Turniere der Männer und Frauen in Peking und noch viel mehr. Viel Spaß bei der nächsten Eishockey-Hörstunde!

In unserer neuesten Podcast-Folge haben Kathrin Fring und Daniel Goldstein unglaublich viel gelernt. Darüber, wie das Eishockey in Tschechien im Allgemeinen und in Litvinov im Speziellen funktioniert. Darüber, wie das Fraueneishockey sich dort entwickelte, darüber wie sich das Eishockey in Salzburg entwickelte und auch darüber, wie Mikael Pancak der wurde, der er heute ist. Pancak ist Nachwuchschef für die jüngsten Spieler*innen beim Heimatclub von Martin und Robert Reichel. Und Mik Pancak ist einer der Hauptcharaktere im Film "Trans I got life".

Mit wem könnten wir gut über Tagebaulandschaften in der Lausitz, die Solidarität innerhalb der DEL und die Weihnachtstage im Kreise der Familie quatschen? Richtig! Stefan Ustorf, Jens Baxmann, Hannes Elster und Daniel Goldstein schwatzen in unserer Weihnachtsausgabe über dies und das und jenes, vor allem aber natürlich über das Eishockey in der DEL und DEL2, in Nürnberg und Weißwasser, bei der U20-WM und bei Olympia und über viele viele andere Dinge. Frohes Fest!

Was ist eigentlich in der DEL2 gerade so los, fragten sich André Rankel und Daniel Goldstein und holten sich zur Beantwortung dieses Punktes Unterstützung vom Tabellenführer aus Ravensburg und den Eispiraten aus Crimmitschau. An beiden Orten spielen ehemalige Kollegen, die einiges zu erzählen hatten. Neben Ravensburg und Crimmitschau spielen auch Bremerhaven, Peiting, Garmisch und Berlin eine Rolle in dieser Folge.

Daniel Goldstein hat schon online mit Küchenmaschinen gehandelt, als Onlinehandel in den Kinderschuhen steckte. Inzwischen gehört sein Shop aus Wetzlar zu den führenden Küchenmaschinenshops in Deutschland. Im Gespräch bringen wir Struktur ins Modellwirrwarr, klären Nutzerfragen und geben Tipps für die praktische Nutzung von Kenwood-Küchenmaschinen zum Brotbacken.

Mit der Kapitänin des deutschen Frauen-Nationalteams und aktuellen Topscorerin der Deutschen Frauen Eishockey Liga (DFEL), Julia Zorn, bekamen Kathrin Fring und Daniel Goldstein in der 49. Sendung von Bend your knees hohen Besuch. Dabei wurde nicht nur, aber natürlich auch über das Olympia-Quali-Turnier gesprochen. Themen außerdem: Das (Nicht-)Comeback von Kathrin Fring, die beeindruckende Saison von Planegg und Julia Zorn, und wie das eigentlich kam, dass sie von der Torhüterin zur Spielerin wurde. Und noch ganz viel mehr. Solltet Ihr eine Meinung haben zum Thema Charity-T-Shirt mit dem Motto "Heraus zum Frauen-Eishockey!", dann meldet Euch doch per Twitter (@dfg193) oder E-Mail dfg19@yahoo.de. Lehitra'ot!

Wir sind im Fieber, im Olympia-Quali-Fieber. Am kommenden Donnerstag geht's für das DEB-Team schon los in Füssen gegen Österreich. Deshalb hört Ihr heute: DIE Vorschau! Verteidigerin Anna-Maria Reich vom ESC Planegg und Stürmerin Hanna Amort von den Eisbären Juniors berichten direkt aus Füssen. Kathrin Fring und Daniel Goldstein lernen etwas mehr über die beiden Frauen, über die Chancen des Teams, sich gegen Italien, Österreich und Dänemark zu qualifizieren und über die unterschiedlichen Karriere-Verläufe einer Landshuterin und einer Lübeckerin. Infos zum Turnier findet Ihr beim DEB: https://bit.ly/Oly-Quali_021121 Die Magenta Sport Doku seht Ihr hier: https://youtu.be/DbP3brfjmdQ

How do we make decisions? What factors influence the final outcome? Do default settings change our preferences? Every decision we make; from choosing something on a menu to deciding whether to be an organ donor, is influenced by our environmental context and the default selections presented to us.  Professor Eric J. Johnson has distilled the latest behavioral science research into his newest book, The Elements of Choice: Why the Way We Decide Matters, and we are thrilled to talk with him about it on this episode of Behavioral Grooves. Eric is Director of the Center for Decision Sciences at Columbia Business School at Columbia University. He examines the interaction between Behavioral Decision Research, Economics and consumer decision making. The implications of his research come together in real world applications such as public policy and marketing. Eric talks with us about how options are presented to decision-makers and how framing affects choices. Our decisions are “assembled”, as Eric likes to put it, in the moment and are not necessarily pre-determined by our preferences. Choice architects have very influential power over decision-makers but Eric highlights to us that we are all designers, and with that comes a moral responsibility.  Topics  (3:56) How Leading Human™ can help with returning to the office. (6:28) Welcome and speed round. (8:00) How interference affects our decision making.  (10:32) The controversy of organ donation defaults. (16:32) We are all designers using the tools of choice architecture. (19:21) How sludge impacts our decision making. (22:42) How context influences defaults. (26:14) What factors moderate the impact of default settings? (29:01) Making choices in the real world vs. classic economic thinking. (32:18) The effects of asking people how long they will live vs. what year they expect to die. (35:04) Smart defaults: defaults set specifically for you. (38:02) What is Query Theory? (39:15) Choice architecture around vaccinations. (42:44) What area does Eric want to research in the future? (44:44) What music does Eric default to? (52:24) Grooving Session with Kurt and Tim on how to apply Eric's research to our lives.   © 2021 Behavioral Grooves Links Behavioral Grooves Patreon: www.patreon.com/behavioralgrooves Leading Human™ Workbook and Playbook: https://www.behavioralgrooves-store.com/products/copy-of-the-leading-human-playbook-workbook-package Leading Human™, Free Whitepaper Download: https://www.behavioralgrooves-store.com/collections/leading-human/products/human-centered-workplace-checklist Leading Human™ Workshop on Dec 14th, 2021 (more dates to be added soon): https://www.behavioralgrooves-store.com/collections/leading-human/products/leading-human-workshop Promo Code: GROOVERS to receive $20 off (limited time offer for listeners). “The Elements of Choice: Why the Way We Decide Matters” by Eric J. Johnson:  https://amzn.to/39yXr20 "Do Defaults Save Lives?" by Johnson, Eric, and Daniel Goldstein. Science 302 (2003): https://www8.gsb.columbia.edu/researcharchive/articles/1275  “Sludge: What Stops Us from Getting Things Done and What to Do about It” by Cass Sunstein: https://amzn.to/3i1lLhD Tim Kachuriak, Episode 221: Donating Our Money Is Irrational, So Why Do We Do It? Tim Kachuriak Explains Our Motivations: https://behavioralgrooves.com/episode/why-we-donate-our-money/ “Time to Retire: Why Americans Claim Benefits Early and How to Encourage Them to Delay” In Behavioral Science and Policy (2015). Coauthor(s): Kirstin Appelt, Melissa Knoll, Eric Johnson, Jonathan Westfall: https://behavioralpolicy.org/articles/time-to-retire-why-americans-claim-benefits-early-how-to-encourage-delay/  Musical Links  Eric Dolphy “Out to Lunch!”: https://www.youtube.com/watch?v=Ne6GCYO8pAc Django Reinhardt “Three-Fingered Lightning”: https://www.youtube.com/watch?v=PQhTpgicdx4  David Grisman Quartet “Dawg Funk”: https://www.youtube.com/watch?v=9P-FQ7xLiso  Jerry Garcia “How Sweet It Is To Be Loved By You”: https://www.youtube.com/watch?v=1_NOFuEb-yo&ab_channel=JerryGarcia  Beethoven: https://www.youtube.com/watch?v=W-fFHeTX70Q  Caroline Shaw & Sō Percussion “Let The Soil Play a Simple Part”: https://www.youtube.com/watch?v=-cHc4n5mgNM&ab_channel=NonesuchRecords  George E. Lewis “Mind In Flux” at the BBC Proms 2021: https://www.youtube.com/watch?v=CSvwisQ3la4  Bruce Springsteen “Tougher Than The Rest”: https://www.youtube.com/watch?v=_91hNV6vuBY 

Es war schon ne Weile her, dass Stefan Ustorf und Daniel Goldstein, allen Bend your knees-Hörer*innen besser bekannt als Usti und Goldi, miteinander gequatscht haben. Währenddessen wurde in Nürnberg der Trainer gewechselt und Stefan stand selbst fünf Spiele an der Bande. Wie es ihm geht und wie die Lage sich derzeit darstellt, das hört Ihr in unserer neuesten Podcast-Ausgabe.

There's a benefit to seeing the law from multiple different perspectives: it gives you an uncanny ability to understand how other people view an issue and what their definition of winning is. Daniel Goldstein credits his suc cess as a lawyer and as a businessperson to his exposure to diverse perspectives. He has seen the world from the perspective of a federal judge as a clerk, from the perspective of a private party hiring a lawyer, from the perspective of the government with the Securities and Exchange Commission, and from the perspective of an in-house lawyer. Currently, he is the EVP and Chief Legal and Compliance Officer at Pitney Bowes , where he leverages the hard-earned wisdom from those different perspectives to advise the business. In this episode, we discuss: - The immense benefit of having multiple perspectives on the law - Managing risk for a 100+ year old company - Partnering effectively with law firms - How the GC role will change over the next 10 years - Learning how to fail Hear more stories by subscribing to Innovative Legal Leadership on Apple Podcasts , Spotify , or any podcast platform. Listening on a desktop & can't see the links? Just search for Innovative Legal Leadership in your favorite podcast player.

In dieser Ausgabe widmen sich André Rankel und Daniel Goldstein ganz und gar einem Ex-Nationalspieler, Ex-DEL-Meister und Stürmer für den SC Riessersee, die Augsburger Panther, den EHC München, die Adler Mannheim, den ERC Ingolstadt, die Eisbären Berlin und die Löwen Frankfurt: Martin Buchwieser. Es geht um seine COVID19-Erkrankung, seine Long-Covid-Symptome und seine Herzmuskelentzündung, die letztendlich zu seinem Karriereende geführt haben, es geht um Kieferbrüche, Watschn-Gesichter und Monchhichis, um schöne Tore, Volksfeste und das Leben in Garmisch-Partenkirchen. Herzlichen Glückwunsch zu einer großen Karriere an Martin Buchwieser sagt Bend your knees und auch Florian Kettemer, Sebastian Elwing und Florian Busch mit Gastauftritten.

There's a benefit to seeing the law from multiple different perspectives: it gives you an uncanny ability to understand how other people view an issue and what their definition of winning is. Daniel Goldstein credits his suc cess as a lawyer and as a businessperson to his exposure to diverse perspectives. He has seen the world from the perspective of a federal judge as a clerk, from the perspective of a private party hiring a lawyer, from the perspective of the government with the Securities and Exchange Commission, and from the perspective of an in-house lawyer. Currently, he is the EVP and Chief Legal and Compliance Officer at Pitney Bowes , where he leverages the hard-earned wisdom from those different perspectives to advise the business. In this episode, we discuss: - The immense benefit of having multiple perspectives on the law - Managing risk for a 100+ year old company - Partnering effectively with law firms - How the GC role will change over the next 10 years - Learning how to fail Hear more stories by subscribing to Innovative Legal Leadership on Apple Podcasts , Spotify , or any podcast platform. Listening on a desktop & can't see the links? Just search for Innovative Legal Leadership in your favorite podcast player.

+++Bully #062+++ Heute haben wir gleich zwei Gäste. Unglaublich, aber wahr. Helmut glänzt heute durch Abwesenheit, aber seid nicht traurig. Wir haben für hochkarätigen Ersatz gesorgt! Zum Ersten haben wir Daniel Goldstein für diese Episode gewinnen können. Das ist ja schon mal eine Meldung wert. Aber dann haben wir noch Yvonne Rothemund vom ESC Planegg in dieser Episode. Und worüber kann man mit den beiden am besten reden? Na klar, Fraueneishockey. Und da gibt es einiges! Die abgelaufenen WM in Calgary, die Frauenbundesliga und die Olympia-Quali im November. Frank lernt in dieser Folge auch, dass es in der Quali nicht vier, sondern drei Spiele sind. Vielen Dank an Yvonne und Daniel! And a very special thanks to "Honeycub" for their Song in this Episode!

Wenn eine promovierte Neuro-Biologin, eine diplomierte Erziehungswissenschaftlerin und diplomierte Immobilienwirtschafterin mit einem Sportjournalisten quatschen, dann ist es Zeit für eine kleine Vorschau auf den Saisonstart der Frauen-Bundesliga (25. September 2021) bei Bend your knees. Die Kapitänin vom KEC "Die Haie" Gesa Dinges und die Ingolstädter Nationalspielerin Bernadette Karpf sind bei Kathrin Fring und Daniel Goldstein zu Gast und sprechen über dies und das und jenes. Viele Informationen und interessante Geschichten zur neuen DFEL-Saison findet Ihr in der neuesten Ausgabe des Eishockeymagazins Dump&Chase. Mehr dazu unter: www.dump-and-chase.com!

Our guest on the podcast today is Professor Hal Hershfield. Dr. Hershfield is Professor of Marketing, Behavioral Decision Making, and Psychology at UCLA's Anderson School of Management. His research concentrates on the psychology of long-term decision-making and how people's perceptions of the passage of time affect the decisions that they make. He has consulted with numerous organizations including Prudential, the Consumer Financial Protection Bureau, Merrill Lynch, and the Principal Financial Group. He received his B.A. at Tufts University and his Ph.D. at Stanford.BackgroundBioThe Pandemic“How Elvis Got Americans to Accept the Polio Vaccine,” by Hal Hershfield and Ilana Brody, scientificamerican.com, Jan. 18, 2021.“Should You Immerse Yourself in Bad News These Days or Ignore It Completely?” by Hal Hershfield, scientificamerican.com, May 5, 2020.“How to Craft the Vaccine Message for the Undecided,” by Bill Kisliuk, ucla.edu, May 4, 2021.“Your Messaging to Older Audiences Is Outdated,” by Hal Hershfield and Laura Carstensen, harvardbusinessreview.org, July 2, 2021.“Time Is Meaningless Now,” by Shayla Love, vice.com, April 10, 2020.Saving for the FutureDaniel Kahneman“Temporal Reframing and Participation in a Savings Program: A Field Experiment,” by Hal Hershfield, Stephen Shu, and Shlomo Benartzi, halhershfield.com, 2020.“Using Vividness Interventions to Improve Financial Decision Making,” by Hal E. Hershfield, Elicia M. John, and Joseph S. Reiff, halhershfield.com, 2018.“Increasing Saving Behavior Through Age-Progressed Renderings of the Future Self,” by Hal Hershfield, Daniel Goldstein, William Sharpe, Jesse Fox, Leo Yeykelis, Laura Carstensen, and Jeremy Bailenson, halhershfield.com, 2011. “The Future Self,” by Hal E. Hershfield and Daniel Bartels, uchicago.edu, 2018.“Beliefs About Whether Spending Implies Wealth,” by Heather Barry Kappes, Joe J. Gladstone, and Hal Hershfield, londonschoolofeconomics.com, 2020.“Seeking Lasting Enjoyment With Limited Money: Financial Constraints Increase Preference for Material Goods Over Experiences,” by Stephanie M. Tully, Hal E. Hershfield, and Tom Meyvis, halhershfield.com, 2015.“Do Images of Older Americans Reinforce Stereotypes?” by Colette Thayer and Laura Skufca, aarp.org, September 2019.Retirement and Older Adults“People Search for Meaning When They Approach a New Decade in Chronological Age,” by Adam Alter and Hal Hershfield, halhershfield.com, 2014.Mike North“You Owe It to Yourself: Boosting Retirement Saving With a Responsibility-Based Appeal,” by Christopher J. Bryan and Hal E. Hershfield, halhershfield.com, 2012.

Unter leicht veränderten Vorzeichen schauen Stefan Ustorf und Daniel Goldstein, jetzt mehr oder weniger in Interviewform, auf den Saisonstart der PENNY DEL. Hört rein, wie der Sportdirektor der Nürnberg Ice Tigers die Vorbereitung seines Teams bewertet, wen er als Gewinner der vergangenen Wochen sieht, wie die Verpflichtung seines Sohnes Jake zustande kam, welchen Neuzugang aus der Liga er gerne in seinem Team gehabt hätte und worauf er sich ganz besonders freut. Und ja, noch viel mehr. BYK, nur echt mit Ustis Meckerecke!

Die Eishockey-WM in Calgary steht vor der Tür und natürlich lässt Euch unser kleiner Eishockeypodcast damit nicht alleine. Bevor die Frauen am 20. August ins Turnier starten erfahrt ihr bei uns alles über die derzeitige Lage in Calgary, das deutsche Team und die Favoritinnen auf den WM-Titel. Mit dabei sind Theresa "Resi" Wagner und Laura Kluge aus dem DEB-Team und Martin Merk von der IIHF. Für die Löcher in ihren Bäuchen zeichnen Kathrin Fring und Daniel Goldstein verantwortlich. Eigentlich aber gehts um ganze andere Themen. Können wir Laura dazu überreden, ein kleines Stück auf der Ukulele zu spielen? Warum ist die WM für Resi die erste überhaupt? Und wer sind die Kabinen-DJs, ersten aufm Eis, Speedy Gonzales, letzten aus Kabine, Schlagschussmonster, Drillbuster, Schlägerverbrauchsköniginnen, Schittschuhschliffpedantinnen, Mannschaftskassenverwalterinnen und Kabinenstimmungsmacherin im DEB-Team? Das BYK-Tippspiel zur WM findet Ihr hier: www.kicktipp.de/byk-fwm21

Tori Murden McClure was the first woman - and the first American - to successfully row across the Atlantic Ocean. She succeeded in 1999 after an attempt in 1998 was foiled by a hurricane. Her vessel? A 23 foot rowboat she had built and named the “American Pearl.” The story of her accomplishment has inspired the new musical “Row” - with a book by Daniel Goldstein and music and lyrics by singer-songwriter, Dawn Landes . The Williamstown Theatre Festival presents the World Premiere Musical “Row” directed by Tyne Rafaeli at The Clark through August 15. Please note, there is limited ticket availability. Dawn Landes and Daniel Goldstein join us, followed by Grace McClean , the actor portraying Tori Murden McClure in "Row." McClean's performances in previous projects have been called “electrifying” by The Huffington Post and “phenomenal” by The New York Times. In addition to performing on Broadway (Natasha, Pierre & The Great Comet of 1812) and Off, McClean makes time for her band Grace

Daniel's advice for the senior generation family members, "Allow the NxtGen to fail."   His advice for the NxtGen, "Be serious about your life and focus on making a difference, this will give you a deep sense of contribution."   Daniel and Francesco have a conversation about how to reduce the danger of entitlement and the burden of being in a family business by establishing an Employee Share Ownership Plan (ESOP).      What are the biggest dangers in a family business? The human beings?

Wir beschließen die erste Staffel Bend your knees mit einem ganz besonderen Gast. Und weil dieser Gast kein normaler ist, wird die Folge etwas anders. Wir erlaubten ihm, seine Muttersprache zu benutzen, weil es ihm leichter fällt in dieser zu quatschen, als in der der Vorfahren seines Vaters. Hört den fünfmaligen Eisbären-Meistergoalie Rob Zepp im Gespräch mit Stefan Ustorf, André Rankel und Daniel Goldstein über seine Familie, über die guten alten und die spannenden neuen Zeiten! Zepp erzählt von seinem neuen Job als Manager für spezielle Projekte bei der nordamerikanischen Eishockey-Spielergewerkschaft NHLPA und erklärt sie uns. Er spricht darüber, welche Position sein Sohn Gavin spielt und welchen Sport seine Tochter macht. Vielen Dank dafür, dass Ihr unsere ersten 36 Ausgaben BYK so zahlreich gehört habt. Wünsche für die Zukunft gerne per E-Mail an dfg19@yahoo.de. Wir danken unserem Presenter TheFanPunktFM und den Unterstützern ROBIN und SpradeTV. Schöne Sommerferien!

Gerade als Ihr dachtet, wir hätten uns bereits ganz heimlich, still und leise in die Sommerpause verabschiedet, da kommen wir mit zwei Ausgaben fürs Ende der ersten Staffel von Bend your knees zurück, die sich gewaschen haben...! Für #35 hatten Kathrin Fring und Daniel Goldstein den Sportmanager der Adler Mannheim, Jan-Axel Alavaara zu Gast. Er erzählt uns, wie die Mannheimer ihr Team für die kommende Saison zusammenstellen, was sie sich von dem einen oder anderen Neuzugang versprechen und wie es mit einem Saisonziel aussieht. Außerdem plaudern wir mit ihm über seine Spielerkarriere, seine Erfahrungen in Göteborg zusammen mit Leuten wie Daniel Alfredsson oder Henrik Lundqvist gespielt zu haben und über das einmalige Erlebnis mit Frölunda nach fast 40 Jahren ohne Titel, Anfang der 2000er zurück an die schwedische Spitze gelangt zu sein. Fahrradcontent inklusive! Am Ende der Folge gibt uns Kathrin noch einen kleinen Einblick in die Sommervorbereitung des DFEL-Teams der Eisbären Juniors und was bei ihr in den kommenden Wochen so ansteht. BYK wird präsentiert von TheFanPunktFM und unterstützt von ROBIN und SpradeTV.

I have come to learn recently that there are two main pathways to happiness - The Hedonic Pathway chasing pleasure and enjoyment while avoiding pain and the Eudaimonic Pathway seeking purpose and meaning through living with virtue. In this episode I attempt to speak to your pain recognizing that you feel trapped in the Hedonic pathway which is causing you suffering in chronic illness, chronic pain, emotional madness, and perhaps caught in a spiritual crisis of identity. The Hedonic Pathway to happiness is fleeting never truly satisfying our Soul. Therefore, pleasure leads to addiction and pain leads to aversion. If you maintain an identity within this pathway then happiness is only found in the people, places and things of this world. Something outside of you makes you happy. You are essentially needing immediate gratification as your 'source' of happiness. Eventually you will feel empty "masturbating with life" and look to fill that emptiness with more Hedonic pleasure thinking more will silence the empty feeling. It doesn't and you don't feel more happiness inside. You have "sold your soul to the devil" It also speaks loudly to the Lord's Prayer Our Father - where it ends with "...and lead us not into temptation but deliver us from evil. Amen." The Eudaimonic Pathway to happiness is eternal because it satisfies the Soul within. It establishes a connection to the source of life and the source of love within oneself. It is not necessarily free from fears or pain, yet it is freedom from suffering as you navigate within yourself, cultivating your true heart to transcend the suffering. Both pathways express the Eastern Philosophy of Karma ~ " What you sow you reap." In Daniel Goldstein's 2011 Ted Talk "The Battle Between Your Present and Future Self" He quotes N.W. Senior as saying "To Obstain from the enjoyment which is in our power, or seek distant rather than immediate results, are among the most painful exertions of the human will." And so living with virtue means cultivating and practicing virtues of Self Mastery, Wisdom, Courage, Love, Hope, Gratitude, Curiosity and Zest. Through which one 'makes love to life' and fill their life with purpose and meaning. To Be and Become the Change in the world you want to see in the world is another quote from Gandhi that hits the Eudaimonic Happiness Pathway bullseye. Here are some links for you to explore and learn more Daniel Goldstein's TedSalon NY2011 Talk https://www.ted.com/talks/daniel_goldstein_the_battle_between_your_present_and_future_self#t-238459 Optimize Me - Optimal Coaching Program to live life with greater Virtue https://www.optimize.me/coach-overview/ For more information about me check out my website https://www.danhegerich.com/ Power and Grace

Dies ist, wie alle anderen BYK-Ausgaben auch, eine ganz besondere. Es ist die mit dem schlechtesten Ton ever, aber einem tollen, weil erfolgreichen Gast. Und einem wieder ganz starken Co-Gastgeber... Jonas Müller nahm sich 24 Stunden vor dem Halbfinale gegen Finnland Zeit für ne halbstündige Sonderausgabe und sprach mit André Rankel und Daniel Goldstein darüber, wie er das Spiel jetzt angeht und natürlich viel viel mehr. Außerdem bekam er Tipps vom WM-Halbfinalisten von 2010. Wenn Ihr aufgrund des schlechten Tons lieber abschaltet, lest den Tagebuch-Eintrag von und mit Jonas beim Tagesspiegel: https://bit.ly/TSP_WM_Blog!

Ihr hattet gedacht die Meista-Ausgabe mit Frank Hördler wäre Eure Lieblingsfolge und könnte nicht getoppt werden? Entscheidet das lieber nach den nächsten 70 Minuten! Hannes Elster und Daniel Goldstein haben die Olympiahelden Leo Pföderl (Stürmer) und André Kreidler (Physio) in der Quarantäne im Hotel im lettischen Riga kurz vor Beginn der Eishockey-WM ins Kreuzverhör genommen. Sehr spannend dabei: Bekommt Leo im Laufe der Sendung sein Mittagessen? Andere zu klärende Fragen: Wie hat er so gut das Bullyspielen gelernt und was war los bei der Meistafeier? Was sind die Aufgaben eines Eishockey-Physios? Wer sind die härtesten Spieler und wie sieht der Kontakt zu anderen Physiotherapeuten aus? BYK wird präsentiert von TheFanPunktFM und unterstützt von SpradeTV sowie ROBIN.

In this episode - Daniel and I discuss the particular benefits of an ESOP to the employees - he does a great job of explaining their ESOP culture at Folience and the remarkable story of integrating other ESOP companies into Folience. This episode also digs into the reality of how ESOPs make a difference for all of our communities!

I welcome Daniel Goldstein, the President and CEO of Folience, to the show. Folience is an investment firm that specializes in ESOPs (Employee Stock Ownership Plans). ESOPs are an intriguing business strategy that every business owner should consider. Daniel explains what ESOPs are and why they are a viable option for exiting founders. He has an impressive professional background and he has a unique perspective on ESOPs and family legacy businesses. Daniel outlines what Folience offers their clients and what they look for in the ideal candidate. What you will learn: Daniel’s banking background. Folience’s history and movement to ESOP. ESOPs are not socialist. How company boards operate in an ESOP. What does a trustee do for the board? The legal side of an ESOP. Folience’s goals for their partners. What does Folience do for their companies? How Folience is different from a private equity firm. How watching company demographics can cultivate new talent. When owners and employee owners get their payouts from the company. E stands for “engagement.” Why ESOPs have to pay fair market value and what that entails. What Folience looks for in a possible company partnership. Folience’s success to date. Daniel’s final thoughts. Takeaway: An ESOP is a serious option for business owners and should be looked into. GEXP Collaborative offers an ultimate guide about ESOPs. We also have a ton of resources for company owners who want to learn more about ESOPs. If you are interested in more information, reach out to me or the website. Links and Resources Daniel’s email Folience GEXP Collaborative GEXP Ultimate Guides About Daniel: Daniel Goldstein is President and CEO of Folience. He has 25 years of executive leadership experience that spans five continents and an irresistible wanderlust that’s carried him to more than 50 countries. It’s a spirit for exploration that’s essential as Daniel shares Folience with the world in his role as chief ambassador and educator. Daniel also guides investment strategy, which includes acquiring, integrating and advising each of the brands in the Folience portfolio.

We often know how to plan for the future, but find it hard to take the necessary steps. This hour, TED speakers challenge the inevitable and explore what to do today to prepare ourselves for tomorrow.TED speakers include neuroscientists Daniel Levitin and Rebecca Brachman, writer (and neuroscientist) Lisa Genova, epidemiologist Seth Berkley, and behavioral economist Daniel Goldstein.