Robust Domain Randomised Reinforcement Learning through Peer-to-Peer Distillation

2020-12-09 · Chenyang Zhao, Timothy Hospedales

In reinforcement learning, domain randomisation is an increasingly popular technique for learning more general policies that are robust to domain-shifts at deployment. However, naively aggregating information from randomised domains may lead to high variance in gradient estimation and unstable learning process. To address this issue, we present a peer-to-peer online distillation strategy for RL termed P2PDRL, where multiple workers are each assigned to a different environment, and exchange knowledge through mutual regularisation based on Kullback-Leibler divergence. Our experiments on continuous control tasks show that P2PDRL enables robust learning across a wider randomisation distribution than baselines, and more robust generalisation to new environments at testing.

📄 PDF Abstract BibTeX arXiv:2012.04839

Code (0)

등록된 구현이 없습니다.

Tasks

continuous-control Continuous Control reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Peer Learning: Learning Complex Policies in Groups from Scratch via Action Recommendations

2023-12-15 · Cedric Derstroff, Mattia Cerrato, Jannis Brugger, Jan Peters 외

Peer learning is a novel high-level reinforcement learning framework for agents learning in groups. While standard reinforcement learning trains an individual agent in trial-and-error fashion, all on its own, peer learni…

OpenAI Gym reinforcement-learning Reinforcement Learning

Reinforcement Learning Enabled Peer-to-Peer Energy Trading for Dairy Farms

2024-05-21 · Mian Ibad Ali Shah, Enda Barrett, Karl Mason

Farm businesses are increasingly adopting renewables to enhance energy efficiency and reduce reliance on fossil fuels and the grid. This shift aims to decrease dairy farms' dependence on traditional electricity grids by …

energy trading

Frustratingly Easy Regularization on Representation Can Boost Deep Reinforcement Learning

2022-05-29 · CVPR 2023 1 · Qiang He, Huangyuan Su, Jieyu Zhang, Xinwen Hou

Deep reinforcement learning (DRL) gives the promise that an agent learns good policy from high-dimensional information, whereas representation learning removes irrelevant and redundant information and retains pertinent i…

Continuous Control Deep Reinforcement Learning reinforcement-learning Reinforcement Learning+2

Transfer Heterogeneous Knowledge Among Peer-to-Peer Teammates: A Model Distillation Approach

2020-02-06 · Zeyue Xue, Shuang Luo, Chao Wu, Pan Zhou 외

Peer-to-peer knowledge transfer in distributed environments has emerged as a promising method since it could accelerate learning and improve team-wide performance without relying on pre-trained teachers in deep reinforce…

Deep Reinforcement Learning Reinforcement Learning Transfer Learning

NLPeer: A Unified Resource for the Computational Study of Peer Review

2022-11-12 · Nils Dycke, Ilia Kuznetsov, Iryna Gurevych

Peer review constitutes a core component of scholarly publishing; yet it demands substantial expertise and training, and is susceptible to errors and biases. Various applications of NLP for peer reviewing assistance aim …