Reward Aware

Reward-aware methods in reinforcement learning aim to optimize agent behavior by effectively incorporating reward signals, whether intrinsic or extrinsic, to guide exploration and policy learning. Current research focuses on improving the efficiency and robustness of reward-guided exploration, particularly in complex environments with high-dimensional observations or multiple objectives, employing techniques like contrastive learning, Bayesian optimization, and multi-view representation learning within various model architectures. These advancements are significant for improving sample efficiency and generalization in reinforcement learning agents, leading to more effective solutions for robotics, control systems, and other applications requiring adaptive decision-making.

Papers

November 5, 2023

Multi-objective Reinforcement Learning with Nonlinear Preferences: Provable Approximation for Maximizing Expected Scalarized Return
Nianli Peng, Muhang Tian, Brandon Fain
Multi Objective Reinforcement Learning Provable Guarantee Risk Aware Reinforcement Learning Multi Objective Markov Decision Process Reward Aware

September 29, 2023

Balancing Both Behavioral Quality and Diversity in Unsupervised Skill Discovery
Xin Liu, Yaran Chen, Dongbin Zhao
Diversity Awareness Skill Discovery Diversity Enhancement Reward Aware Multi Objective Reward

April 14, 2023

Towards Controllable Diffusion Models via Reward-Guided Exploration
Hengtong Zhang, Tingyang Xu
Diffusion Model Policy Gradient Conditional Diffusion Model Controllable Diffusion Model Reward Aware

March 17, 2023

Optimal Horizon-Free Reward-Free Exploration for Linear Mixture MDPs
Junkai Zhang, Weitong Zhang, Quanquan Gu
Reward Free Reward Free Reinforcement Learning Reward Free Exploration Linear Mixture Markov Decision Process Reward Aware

January 19, 2023

Keyframe Demonstration Seeded and Bayesian Optimized Policy Search
Onur Berk Tore, Farzin Negahbani, Baris Akgun
Robot Skill Demonstration Framework Reward Aware

November 28, 2022

Tackling Visual Control via Multi-View Exploration Maximization
Mingqi Yuan, Xin Jin, Bo Li, Wenjun Zeng
Reinforcement Learning Multi View Representation Learning Visual Control Reward Aware

June 21, 2022

On the Statistical Efficiency of Reward-Free Exploration in Non-Linear RL
Jinglin Chen, Aditya Modi, Akshay Krishnamurthy, Nan Jiang, Alekh Agarwal
Low Rank Sample Efficiency Reward Free Reinforcement Learning Reward Free Exploration Reward Aware

May 22, 2022

Near-Optimal Algorithms for Autonomous Exploration and Multi-Goal Stochastic Shortest Path
Haoyuan Cai, Tengyu Ma, Simon Du
Autonomous Exploration Goal Conditioned Policy Near Optimal Algorithm Stochastic Shortest Path Reward Aware

December 1, 2021

DOPE: Doubly Optimistic and Pessimistic Exploration for Safe Reinforcement Learning
Archana Bura, Aria HasanzadeZonuzy, Dileep Kalathil, Srinivas Shakkottai, Jean-Francois Chamberland
Safe Reinforcement Learning Finite Horizon Simple Regret Optimistic Exploration Model Based Safe Reinforcement Learning Reward Aware