Meta Reinforcement Learning

Meta-reinforcement learning (Meta-RL) aims to create agents capable of rapidly adapting to new tasks with minimal experience, leveraging prior learning to accelerate adaptation. Current research focuses on improving sample efficiency and generalization across diverse tasks, employing model architectures like recurrent neural networks, transformers, and hypernetworks, along with algorithms such as MAML and PPO, often incorporating techniques like importance sampling and constrained optimization. This field is significant for its potential to enable more robust and adaptable AI systems in various applications, from resource optimization in wireless networks to safe and efficient robot control in unpredictable environments.

Papers

May 20, 2023

On First-Order Meta-Reinforcement Learning with Moreau Envelopes
Mohammad Taha Toghani, Sebastian Perez-Salazar, César A. Uribe
Policy Gradient Policy OpTimization Meta Reinforcement Learning Moreau Envelope

April 29, 2023

Meta-Reinforcement Learning Based on Self-Supervised Task Representation Learning
Mingyang Wang, Zhenshan Bing, Xiangtong Yao, Shuai Wang, Hang Su, Chenguang Yang, Kai Huang, Alois Knoll
Self Supervised Meta Reinforcement Learning Task Distribution Meta Reinforcement Robust Task

February 15, 2023

Meta-Reinforcement Learning via Exploratory Task Clustering
Zhendong Chu, Hongning Wang
Meta Reinforcement Learning Task Distribution Exploration Policy Sample Efficient Policy Task Clustering

February 11, 2023

Procedural generation of meta-reinforcement learning tasks
Thomas Miconi
Meta Reinforcement Learning Procedural Content Generation Meta Training

January 26, 2023

Train Hard, Fight Easy: Robust Meta Reinforcement Learning
Ido Greenberg, Shie Mannor, Gal Chechik, Eli Meirom
Reinforcement Learning Meta Reinforcement Learning High Speed Train System Robust Meta Reinforcement Learning Robust Objective

January 19, 2023

A Survey of Meta-Reinforcement Learning
Jacob Beck, Risto Vuorio, Evan Zheran Liu, Zheng Xiong, Luisa Zintgraf, Chelsea Finn, Shimon Whiteson
Timely Survey Deep Reinforcement Learning Meta Reinforcement Learning

January 15, 2023

Neuro-symbolic Meta Reinforcement Learning for Trading
S I Harini, Gautam Shroff, Ashwin Srinivasan, Prayushi Faldu, Lovekesh Vig
Reinforcement Learning Meta Reinforcement Learning Sequential Decision Making Problem Financial Trading Symbolic Learning

December 30, 2022

POMRL: No-Regret Learning-to-Plan with Increasing Horizons
Khimya Khetarpal, Claire Vernade, Brendan O'Donoghue, Satinder Singh, Tom Zahavy
Model Uncertainty Meta Reinforcement Learning Single Task Planning Loss New Horizon Planning Horizon

December 17, 2022

Cognitive Level-$k$ Meta-Learning for Safe and Pedestrian-Aware Autonomous Driving
Haozhe Lei, Quanyan Zhu
Autonomous Driving Meta Reinforcement Learning Driving Algorithm Interaction Aware

November 20, 2022

Efficient Meta Reinforcement Learning for Preference-based Fast Adaptation
Zhizhou Ren, Anji Liu, Yitao Liang, Jian Peng, Jianzhu Ma
Meta Reinforcement Learning Fast Adaptation Reinforcement Learning Benchmark Policy Adaptation Human in the Loop Reinforcement Meta Rl Algorithm

November 19, 2022

Debiasing Meta-Gradient Reinforcement Learning by Learning the Outer Value Function
Clément Bonnet, Laurence Midgley, Alexandre Laterre
Value Function Meta Reinforcement Learning Meta Gradient Discount Factor

October 29, 2022

BIMRL: Brain Inspired Meta Reinforcement Learning
Seyed Roozbeh Razavi Rohani, Saeed Hedayatian, Mahdieh Soleymani Baghshah
Reinforcement Learning Brain Function Meta Reinforcement Learning Intrinsic Reward

October 27, 2022

Meta-Reinforcement Learning Using Model Parameters
Gabriel Hartmann, Amos Azaria
Meta Reinforcement Learning Model Parameter Environment Dynamic

October 23, 2022

MetaEMS: A Meta Reinforcement Learning-based Control Framework for Building Energy Management System
Huiliang Zhang, Di Wu, Benoit Boulet
Reinforcement Learning Meta Reinforcement Learning Building Energy Energy Management System

October 20, 2022

Hypernetworks in Meta-Reinforcement Learning
Jacob Beck, Matthew Thomas Jackson, Risto Vuorio, Shimon Whiteson
Reinforcement Learning Multi Task Meta Reinforcement Learning Attention Hypernetworks Robotics Benchmark

October 11, 2022

Meta-Learning with Self-Improving Momentum Target
Jihoon Tack, Jongjin Park, Hankook Lee, Jaeho Lee, Jinwoo Shin
Meta Reinforcement Learning Meta Training Adaptive Momentum

October 9, 2022

Decomposed Mutual Information Optimization for Generalized Context in Meta-Reinforcement Learning
Yao Mu, Yuzheng Zhuang, Fei Ni, Bin Wang, Jianyu Chen, Jianye Hao, Ping Luo
Context Learning Mutual Information Meta Reinforcement Learning Transition Dynamic Dynamic Generalization

October 7, 2022

Robotic Control Using Model Based Meta Adaption
Karam Daaboul, Joel Ikels, Marius Zöllner
Reinforcement Learning Robot Control Meta Reinforcement Learning Model Based Adaptation

October 6, 2022