Reward Maximizing Policy

Reward-maximizing policies aim to design agents or systems that optimally select actions to achieve the highest possible reward, as defined by a specific objective function. Current research focuses on improving the efficiency and stability of training these policies, particularly within the context of large language models and reinforcement learning, employing techniques like value-augmented sampling and constrained policy optimization to address challenges such as safety and generalization across diverse tasks. These advancements are significant for developing more robust and adaptable AI systems, with applications ranging from personalized language models to efficient robotic control.

Papers

October 17, 2024

ORSO: Accelerating Reward Design via Online Reward Selection and Policy Optimization
Chen Bo Calvin Zhang, Zhang-Wei Hong, Aldo Pacchiano, Pulkit Agrawal
Optimization Purpose Reward Function Reward Shaping Health Policy Recommendation Reward Design Reward Maximizing Policy

May 10, 2024

Value Augmented Sampling for Language Model Alignment and Personalization
Seungwook Han, Idan Shenfeld, Akash Srivastava, Yoon Kim, Pulkit Agrawal
Reinforcement Learning Reward Maximization Language Model Alignment LLM Adaptation Augmented Sample Reward Maximizing Policy

April 17, 2024

Stepwise Alignment for Constrained Language Model Policy Optimization
Akifumi Wachi, Thien Q. Tran, Rei Sato, Takumi Tanabe, Youhei Akimoto
Language Model Direct Preference Optimization Constrained Policy Optimization Bidirectional Alignment Reward Maximizing Policy

March 9, 2023

Task Aware Dreamer for Task Generalization in Reinforcement Learning
Chengyang Ying, Zhongkai Hao, Xinning Zhou, Hang Su, Songming Liu, Dong Yan, Jun Zhu
Reinforcement Learning Task Generalization Reward Maximizing Policy

October 15, 2022

A Policy-Guided Imitation Approach for Offline Reinforcement Learning
Haoran Xu, Li Jiang, Jianxiong Li, Xianyuan Zhan
Offline Reinforcement Learning Based Imitation State of the Art Imitation Imitation Learned Policy Reward Maximizing Policy

October 11, 2022

Mastering the Game of No-Press Diplomacy via Human-Regularized Reinforcement Learning and Planning
Anton Bakhtin, David J Wu, Adam Lerer, Jonathan Gray, Athul Paul Jacob, Gabriele Farina, Alexander H Miller, Noam Brown
Multi Agent Task Planning Game Content Regularized Reinforcement Learning Imitation Learned Policy Self Play Reinforcement Learning Reward Maximizing Policy

October 6, 2022

Exploration via Planning for Information about the Optimal Trajectory
Viraj Mehta, Ian Char, Joseph Abbate, Rory Conlin, Mark D. Boyer, Stefano Ermon, Jeff Schneider, Willie Neiswanger
Reinforcement Learning Task Planning Environment Exploration Full Information Optimal Trajectory Exploration Method Real World Control Reward Maximizing Policy

November 17, 2021

GFlowNet Foundations
Yoshua Bengio, Salem Lahlou, Tristan Deleu, Edward J. Hu, Mo Tiwari, Emmanuel Bengio
Generative Approach Generative Flow Network Energy Based GFlowNets Joint Distribution Reward Maximizing Policy