CURL: Contrastive Unsupervised Representations for Reinforcement Learning
Contrastive Learning has been an established method in NLP and Image classification. The authors show that with relatively minor adjustments, CL can be used to augment and improve RL dramatically.
Abstract: We present CURL: Contrastive Unsupervised Representations for Reinforcement Learning. CURL extracts high-level features from raw pixels using contrastive learning and performs off-policy control on top of the extracted features. CURL outperforms prior pixel-based methods, both model-based and model-free, on complex tasks in the DeepMind Control Suite and Atari Games showing 2.8x and 1.6x performance gains respectively at the 100K interaction steps benchmark. On the DeepMind Control Suite, CURL is the first image-based algorithm to nearly match the sample-efficiency and performance of methods that use state-based features.
Authors: Aravind Srinivas, Michael Laskin, Pieter Abbeel
#jepa #ai #machinelearning
Yann LeCun's position paper on a path towards machine intelligence combines Self-Supervised Learning, Energy-Based Models, and hierarchical predictive embedding models to arrive at a system that can teach itself to learn useful abstractions at multiple levels and use that as a world model to plan ahead in time.
OUTLINE:
0:00 - Introduction
2:00 - Main Contributions
5:45 - Mode 1 and Mode 2 actors
15:40 - Self-Supervised Learning and Energy-Based Models
20:15 - Introducing latent variables
25:00 - The problem of collapse
29:50 - Contrastive vs regularized methods
36:00 - The JEPA architecture
47:00 - Hierarchical JEPA (H-JEPA)
53:00 - Broader relevance
56:00 - Summary & Comments
Paper: https://openreview.net/forum?id=BZ5a1r-kVsf
Abstract: How could machines learn as efficiently as humans and animals? How could machines learn to reason and plan? How could machines learn representations of percepts and action plans at multiple levels of abstraction, enabling them to reason, predict, and plan at multiple time horizons? This position paper proposes an architecture and training paradigms with which to construct autonomous intelligent agents. It combines concepts such as configurable predictive world model, behavior driven through intrinsic motivation, and hierarchical joint embedding architectures trained with self-supervised learning.
Author: Yann LeCun
Links:
Homepage: https://ykilcher.com
Merch: https://ykilcher.com/merch
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://ykilcher.com/discord
LinkedIn: https://www.linkedin.com/in/ykilcher
If you want to support me, the best thing to do is to share out the content :)
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar: https://www.subscribestar.com/yannickilcher
Patreon: https://www.patreon.com/yannickilcher
Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq
Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2
Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m
Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n
...
https://www.youtube.com/watch?v=jSdHmImyUjk
#genderbias #algorithmicfairness #debiasing
A brief look into gender stereotypes in Google Translate. The origin is a Tweet containing a Hungarian text. Hungarian is a gender-neutral language, so translating gender pronouns is ambiguous. Turns out that Google Translate assigns very stereotypical pronouns. In this video, we'll have a look at the origins and possible solutions to this problem.
OUTLINE:
0:00 - Intro
1:10 - Digging Deeper
2:30 - How does Machine Translation work?
3:50 - Training Data Problems
4:40 - Learning Algorithm Problems
5:45 - Argmax Output Problems
6:45 - Pragmatics
7:50 - More on Google Translate
9:40 - Social Engineering
11:15 - Conclusion
Songs:
Like That - Anno Domini Beats
Submarine - Dyalla
Dude - Patrick Patrikios
Links:
TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://discord.gg/4H8xxDF
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Parler: https://parler.com/profile/YannicKilcher
LinkedIn: https://www.linkedin.com/in/yannic-kilcher-488534136/
BiliBili: https://space.bilibili.com/1824646584
If you want to support me, the best thing to do is to share out the content :)
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar: https://www.subscribestar.com/yannickilcher
Patreon: https://www.patreon.com/yannickilcher
Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq
Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2
Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m
Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n
...
https://www.youtube.com/watch?v=J7CrtblmMnU
#metarim #deeprl #catastrophicforgetting
Reinforcement Learning is very tricky in environments where the objective shifts over time. This paper explores agents in multi-task environments that are usually subject to catastrophic forgetting. Building on the concept of Recurrent Independent Mechanisms (RIM), the authors propose to separate the learning procedures for the mechanism parameters (fast) and the attention parameters (slow) and achieve superior results and more stability, and even better zero-shot transfer performance.
OUTLINE:
0:00 - Intro & Overview
3:30 - Recombining pieces of knowledge
11:30 - Controllers as recurrent neural networks
14:20 - Recurrent Independent Mechanisms
21:20 - Learning at different time scales
28:40 - Experimental Results & My Criticism
44:20 - Conclusion & Comments
Paper: https://arxiv.org/abs/2105.08710
RIM Paper: https://arxiv.org/abs/1909.10893
Abstract:
Decomposing knowledge into interchangeable pieces promises a generalization advantage when there are changes in distribution. A learning agent interacting with its environment is likely to be faced with situations requiring novel combinations of existing pieces of knowledge. We hypothesize that such a decomposition of knowledge is particularly relevant for being able to generalize in a systematic manner to out-of-distribution changes. To study these ideas, we propose a particular training framework in which we assume that the pieces of knowledge an agent needs and its reward function are stationary and can be re-used across tasks. An attention mechanism dynamically selects which modules can be adapted to the current task, and the parameters of the selected modules are allowed to change quickly as the learner is confronted with variations in what it experiences, while the parameters of the attention mechanisms act as stable, slowly changing, meta-parameters. We focus on pieces of knowledge captured by an ensemble of modules sparsely communicating with each other via a bottleneck of attention. We find that meta-learning the modular aspects of the proposed system greatly helps in achieving faster adaptation in a reinforcement learning setup involving navigation in a partially observed grid world with image-level input. We also find that reversing the role of parameters and meta-parameters does not work nearly as well, suggesting a particular role for fast adaptation of the dynamically selected modules.
Authors: Kanika Madan, Nan Rosemary Ke, Anirudh Goyal, Bernhard Schölkopf, Yoshua Bengio
Links:
TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://discord.gg/4H8xxDF
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.mi
...
https://www.youtube.com/watch?v=kU-tWy_wr78
#hypertransformer #metalearning #deeplearning
This video contains a paper explanation and an interview with author Andrey Zhmoginov!
Few-shot learning is an interesting sub-field in meta-learning, with wide applications, such as creating personalized models based on just a handful of data points. Traditionally, approaches have followed the BERT approach where a large model is pre-trained and then fine-tuned. However, this couples the size of the final model to the size of the model that has been pre-trained. Similar problems exist with "true" meta-learners, such as MaML. HyperTransformer fundamentally decouples the meta-learner from the size of the final model by directly predicting the weights of the final model. The HyperTransformer takes the few-shot dataset as a whole into its context and predicts either one or multiple layers of a (small) ConvNet, meaning its output are the weights of the convolution filters. Interestingly, and with the correct engineering care, this actually appears to deliver promising results and can be extended in many ways.
OUTLINE:
0:00 - Intro & Overview
3:05 - Weight-generation vs Fine-tuning for few-shot learning
10:10 - HyperTransformer model architecture overview
22:30 - Why the self-attention mechanism is useful here
34:45 - Start of Interview
39:45 - Can neural networks even produce weights of other networks?
47:00 - How complex does the computational graph get?
49:45 - Why are transformers particularly good here?
58:30 - What can the attention maps tell us about the algorithm?
1:07:00 - How could we produce larger weights?
1:09:30 - Diving into experimental results
1:14:30 - What questions remain open?
Paper: https://arxiv.org/abs/2201.04182
ERRATA: I introduce Max Vladymyrov as Mark Vladymyrov
Abstract:
In this work we propose a HyperTransformer, a transformer-based model for few-shot learning that generates weights of a convolutional neural network (CNN) directly from support samples. Since the dependence of a small generated CNN model on a specific task is encoded by a high-capacity transformer model, we effectively decouple the complexity of the large task space from the complexity of individual tasks. Our method is particularly effective for small target CNN architectures where learning a fixed universal task-independent embedding is not optimal and better performance is attained when the information about the task can modulate all model parameters. For larger models we discover that generating the last layer alone allows us to produce competitive or better results than those obtained with state-of-the-art methods while being end-to-end differentiable. Finally, we extend our approach to a semi-supervised regime utilizing unlabeled samples in the support set and further improving few-shot performance.
Aut
...
https://www.youtube.com/watch?v=D6osiiEoV0w
The imputer is a sequence-to-sequence model that strikes a balance between fully autoregressive models with long inference times and fully non-autoregressive models with fast inference. The imputer achieves constant decoding time independent of sequence length by exploiting dynamic programming.
https://arxiv.org/abs/2002.08926
Abstract:
This paper presents the Imputer, a neural sequence model that generates output sequences iteratively via imputations. The Imputer is an iterative generative model, requiring only a constant number of generation steps independent of the number of input or output tokens. The Imputer can be trained to approximately marginalize over all possible alignments between the input and output sequences, and all possible generation orders. We present a tractable dynamic programming training algorithm, which yields a lower bound on the log marginal likelihood. When applied to end-to-end speech recognition, the Imputer outperforms prior non-autoregressive models and achieves competitive results to autoregressive models. On LibriSpeech test-other, the Imputer achieves 11.1 WER, outperforming CTC at 13.0 WER and seq2seq at 12.5 WER.
Authors: William Chan, Chitwan Saharia, Geoffrey Hinton, Mohammad Norouzi, Navdeep Jaitly
Links:
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
...
https://www.youtube.com/watch?v=AU30czb4iQA
One CNN to rule them all! BiT is a pre-trained ResNet that can be used as a starting point for any visual task. This paper explains what it takes to pre-train such a large model and details how fine-tuning on downstream tasks is done best.
Paper: https://arxiv.org/abs/1912.11370
Code & Models: TBA
Abstract:
Transfer of pre-trained representations improves sample efficiency and simplifies hyperparameter tuning when training deep neural networks for vision. We revisit the paradigm of pre-training on large supervised datasets and fine-tuning the model on a target task. We scale up pre-training, and propose a simple recipe that we call Big Transfer (BiT). By combining a few carefully selected components, and transferring using a simple heuristic, we achieve strong performance on over 20 datasets. BiT performs well across a surprisingly wide range of data regimes -- from 1 example per class to 1M total examples. BiT achieves 87.5% top-1 accuracy on ILSVRC-2012, 99.4% on CIFAR-10, and 76.3% on the 19 task Visual Task Adaptation Benchmark (VTAB). On small datasets, BiT attains 76.8% on ILSVRC-2012 with 10 examples per class, and 97.0% on CIFAR-10 with 10 examples per class. We conduct detailed analysis of the main components that lead to high transfer performance.
Authors: Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai, Joan Puigcerver, Jessica Yung, Sylvain Gelly, Neil Houlsby
Links:
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
...
https://www.youtube.com/watch?v=k1GOF2jmX7c
#vjepa #meta #unsupervisedlearning
V-JEPA is a method for unsupervised representation learning of video data by using only latent representation prediction as objective function.
Weights & Biases course on Structured LLM Outputs: https://wandb.me/course-yannic
OUTLINE:
0:00 - Intro
1:45 - Predictive Feature Principle
8:00 - Weights & Biases course on Structured LLM Outputs
9:45 - The original JEPA architecture
27:30 - V-JEPA Concept
33:15 - V-JEPA Architecture
44:30 - Experimental Results
46:30 - Qualitative Evaluation via Decoding
Blog: https://ai.meta.com/blog/v-jepa-yann-lecun-ai-model-video-joint-embedding-predictive-architecture/
Paper: https://ai.meta.com/research/publications/revisiting-feature-prediction-for-learning-visual-representations-from-video/
Abstract:
This paper explores feature prediction as a stand-alone objective for unsupervised learning from video and introduces V-JEPA, a collection of vision models trained solely using a feature prediction objective, without the use of pretrained image encoders, text, negative examples, reconstruction, or other sources of supervision. The models are trained on 2 million videos collected from public datasets and are evaluated on downstream image and video tasks. Our results show that learning by predicting video features leads to versatile visual representations that perform well on both motion and appearance-based tasks, without adaption of the model’s parameters; e.g., using a frozen backbone, our largest model, a ViT-H/16 trained only on videos, obtains 81.9% on Kinetics-400, 72.2% on Something-Something-v2, and 77.9% on ImageNet1K.
Authors: Adrien Bardes Quentin Garrido Xinlei Chen Michael Rabbat Yann LeCun Mido Assran Nicolas Ballas Jean Ponce
Links:
Homepage: https://ykilcher.com
Merch: https://ykilcher.com/merch
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://ykilcher.com/discord
LinkedIn: https://www.linkedin.com/in/ykilcher
If you want to support me, the best thing to do is to share out the content :)
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar: https://www.subscribestar.com/yannickilcher
Patreon: https://www.patreon.com/yannickilcher
Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq
Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2
Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m
Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n
...
https://www.youtube.com/watch?v=7UkJPwz_N_0
#minecraft #neuralnetwork #backpropagation
I built an analog neural network in vanilla Minecraft without any mods or command blocks. The network uses Redstone wire power strengths to carry the signal through one hidden layer, including nonlinearities, and then do automatic backpropagation and even weight updates.
OUTLINE:
0:00 - Intro & Overview
1:50 - Redstone Components Explained
5:00 - Analog Multiplication in Redstone
7:00 - Gradient Descent for Square Root Computation
9:35 - Neural Network Demonstration
10:45 - Network Schema Explained
18:35 - The Network Learns a Datapoint
20:20 - Outro & Conclusion
I built this during a series of live streams and want to thank everyone who helped me and cheered for me in the chat!
World saves here: https://github.com/yk/minecraft-neural-network
Game here: https://www.minecraft.net
Multiplier Inspiration: https://www.youtube.com/channel/UCLmzk4TlnLXCXCHcjuJe2ag
Credits to Lanz for editing!
Links:
TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick
YouTube: https://www.youtube.com/c/yannickilcher
Twitter: https://twitter.com/ykilcher
Discord: https://discord.gg/4H8xxDF
BitChute: https://www.bitchute.com/channel/yannic-kilcher
Minds: https://www.minds.com/ykilcher
Parler: https://parler.com/profile/YannicKilcher
LinkedIn: https://www.linkedin.com/in/yannic-kilcher-488534136/
BiliBili: https://space.bilibili.com/1824646584
If you want to support me, the best thing to do is to share out the content :)
If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this):
SubscribeStar: https://www.subscribestar.com/yannickilcher
Patreon: https://www.patreon.com/yannickilcher
Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq
Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2
Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m
Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n
...
https://www.youtube.com/watch?v=7OdhtAiPfWY