paper-with-me

홈 › Papers

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

2025-03-19 · Kevin Wang, Ishaan Javali, Michał Bortkiewicz, Tomasz Trzciński, Benjamin Eysenbach

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.

📄 PDF Abstract BibTeX arXiv:2503.14858

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning (RL)Self-Supervised Learning

Methods 이 논문이 사용한 방법론

Air Canada Customer Service Number 설명 없음

Similar Papers 제목 키워드 기반

DepthFL: Depthwise Federated Learning for Heterogeneous Clients

2023-05-01 · ICLR 2023 5 · Minjae Kim, Sangyoon Yu, Suhyun Kim, Soo-Mook Moon

Federated learning is for training a global model without collecting private local data from clients. As they repeatedly need to upload locally-updated weights or gradients instead, clients require both computation and c…

Federated Learning

The Curse of Depth in Large Language Models

2025-02-09 · Wenfang Sun, Xinyuan Song, Pengxiang Li, Lu Yin 외

In this paper, we introduce the Curse of Depth, a concept that highlights, explains, and addresses the recent observation in modern Large Language Models(LLMs) where nearly half of the layers are less effective than expe…

Calibrating Self-supervised Monocular Depth Estimation

2020-09-16 · Robert McCraith, Lukas Neumann, Andrea Vedaldi

In the recent years, many methods demonstrated the ability of neural networks to learn depth and pose changes in a sequence of images, using only self-supervision as the training signal. Whilst the networks achieve good …

Depth EstimationMonocular Depth Estimation

RealMonoDepth: Self-Supervised Monocular Depth Estimation for General Scenes

2020-04-14 · Mertalp Ocal, Armin Mustafa

We present a generalised self-supervised learning approach for monocular estimation of the real depth across scenes with diverse depth ranges from 1--100s of meters. Existing supervised methods for monocular depth estima…

Depth EstimationMonocular Depth EstimationSelf-Supervised Learning

Progressive Depth Up-scaling via Optimal Transport

2025-08-11 · Mingzi Cao, Xi Wang, Nikolaos Aletras arxiv

Scaling Large Language Models (LLMs) yields performance gains but incurs substantial training costs. Depth up-scaling offers training efficiency by adding new layers to pre-trained models. However, most existing methods …