1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities

• URL: http://arxiv.org/abs/2503.14858v2
• Date: Sat, 22 Mar 2025 22:24:37 GMT
• Title: 1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities
• Authors: Kevin Wang, Ishaan Javali, Michał Bortkiewicz, Tomasz Trzciński, Benjamin Eysenbach,
• Abstract summary: Scaling up self-supervised learning has driven breakthroughs in language and vision, yet comparable progress has remained elusive in reinforcement learning (RL)<n>In this paper, we study building blocks for self-supervised RL that unlock substantial improvements in scalability, with network depth serving as a critical factor.
• Score: 18.997261006638798
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Scaling up self-supervised learning has driven breakthroughs in language and vision, yet comparable progress has remained elusive in reinforcement learning (RL). In this paper, we study building blocks for self-supervised RL that unlock substantial improvements in scalability, with network depth serving as a critical factor. Whereas most RL papers in recent years have relied on shallow architectures (around 2 - 5 layers), we demonstrate that increasing the depth up to 1024 layers can significantly boost performance. Our experiments are conducted in an unsupervised goal-conditioned setting, where no demonstrations or rewards are provided, so an agent must explore (from scratch) and learn how to maximize the likelihood of reaching commanded goals. Evaluated on simulated locomotion and manipulation tasks, our approach increases performance by $2\times$ - $50\times$. Increasing the model depth not only increases success rates but also qualitatively changes the behaviors learned.

Related papers

• GTR: Guided Thought Reinforcement Prevents Thought Collapse in RL-based VLM Agent Training [62.536191233049614]
  Reinforcement learning with verifiable outcome rewards (RLVR) has effectively scaled up chain-of-thought (CoT) reasoning in large language models (LLMs)<n>This work investigates this problem through extensive experiments on complex card games, such as 24 points, and embodied tasks from ALFWorld.<n>We find that when rewards are based solely on action outcomes, RL fails to incentivize CoT reasoning in VLMs, instead leading to a phenomenon we termed thought collapse.
  arXiv  Detail & Related papers  (2025-03-11T15:17:02Z)
• Accelerating Goal-Conditioned RL Algorithms and Research [17.155006770675904]
  Self-supervised goal-conditioned reinforcement learning (GCRL) agents discover new behaviors by learning from the goals achieved during unstructured interaction with the environment. These methods have failed to see similar success due to a lack of data from slow environment simulations as well as a lack of stable algorithms. We release a benchmark (JaxGCRL) for self-supervised GCRL, enabling researchers to train agents for millions of environment steps in minutes on a single GPU.
  arXiv  Detail & Related papers  (2024-08-20T17:58:40Z)
• Affordance-Guided Reinforcement Learning via Visual Prompting [51.361977466993345]
  Keypoint-based Affordance Guidance for Improvements (KAGI) is a method leveraging rewards shaped by vision-language models (VLMs) for autonomous RL.<n>On real-world manipulation tasks specified by natural language descriptions, KAGI improves the sample efficiency of autonomous RL and enables successful task completion in 30K online fine-tuning steps.
  arXiv  Detail & Related papers  (2024-07-14T21:41:29Z)
• Stop Regressing: Training Value Functions via Classification for Scalable Deep RL [109.44370201929246]
  We show that training value functions with categorical cross-entropy improves performance and scalability in a variety of domains. These include: single-task RL on Atari 2600 games with SoftMoEs, multi-task RL on Atari with large-scale ResNets, robotic manipulation with Q-transformers, playing Chess without search, and a language-agent Wordle task with high-capacity Transformers.
  arXiv  Detail & Related papers  (2024-03-06T18:55:47Z)
• Intelligent gradient amplification for deep neural networks [2.610003394404622]
  In particular, deep learning models require larger training times as the depth of a model increases. Several solutions address these problems independently, but there have been minimal efforts to identify an integrated solution. In this work, we intelligently determine which layers of a deep learning model to apply gradient amplification to, using a formulated approach.
  arXiv  Detail & Related papers  (2023-05-29T03:38:09Z)
• Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels [112.63440666617494]
  Reinforcement learning algorithms can succeed but require large amounts of interactions between the agent and the environment. We propose a new method to solve it, using unsupervised model-based RL, for pre-training the agent. We show robust performance on the Real-Word RL benchmark, hinting at resiliency to environment perturbations during adaptation.
  arXiv  Detail & Related papers  (2022-09-24T14:22:29Z)
• Model-Free Generative Replay for Lifelong Reinforcement Learning: Application to Starcraft-2 [5.239932780277599]
  Generative replay (GR) is a biologically-inspired replay mechanism that augments learning experiences with self-labelled examples. We present a version of GR for LRL that satisfies two desideratas: (a) Introspective density modelling of the latent representations of policies learned using deep RL, and (b) Model-free end-to-end learning.
  arXiv  Detail & Related papers  (2022-08-09T22:00:28Z)
• FOSTER: Feature Boosting and Compression for Class-Incremental Learning [52.603520403933985]
  Deep neural networks suffer from catastrophic forgetting when learning new categories. We propose a novel two-stage learning paradigm FOSTER, empowering the model to learn new categories adaptively.
  arXiv  Detail & Related papers  (2022-04-10T11:38:33Z)
• Improving Computational Efficiency in Visual Reinforcement Learning via Stored Embeddings [89.63764845984076]
  We present Stored Embeddings for Efficient Reinforcement Learning (SEER) SEER is a simple modification of existing off-policy deep reinforcement learning methods. We show that SEER does not degrade the performance of RLizable agents while significantly saving computation and memory.
  arXiv  Detail & Related papers  (2021-03-04T08:14:10Z)
• A Layer-Wise Information Reinforcement Approach to Improve Learning in Deep Belief Networks [0.4893345190925178]
  This paper proposes the Residual Deep Belief Network, which considers the information reinforcement layer-by-layer to improve the feature extraction and knowledge retaining. Experiments conducted over three public datasets demonstrate its robustness concerning the task of binary image classification.
  arXiv  Detail & Related papers  (2021-01-17T18:53:18Z)
• Uniform State Abstraction For Reinforcement Learning [6.624726878647541]
  MultiGrid Reinforcement Learning (MRL) has shown that abstract knowledge in the form of a potential function can be learned almost solely from agent interaction with the environment. In this paper we extend and improve MRL to take advantage of modern Deep Learning algorithms such as Deep Q-Networks (DQN)
  arXiv  Detail & Related papers  (2020-04-06T18:13:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.