Reinforcement Learning

Reinforcement learning (RL) focuses on training agents to make optimal decisions in an environment by learning through trial and error, aiming to maximize cumulative rewards. Current research emphasizes improving RL's efficiency and robustness, particularly in areas like human-in-the-loop training (e.g., using human feedback to refine models), handling uncertainty and sparse rewards, and scaling to complex tasks (e.g., robotics, autonomous driving). Prominent approaches involve various policy gradient methods, Monte Carlo Tree Search, and the integration of large language models for improved decision-making and task decomposition. These advancements are driving progress in diverse fields, including robotics, game playing, and the development of more human-aligned AI systems.

Papers

July 2, 2024

Reinforcement Learning from Human Feedback: Whose Culture, Whose Values, Whose Perspectives?
Kristian González Barman, Simon Lohse, Henk de Regt
Reinforcement Learning Human Feedback Synthesized View Policy Value Local Culture Pluralistic Alignment
Text-Aware Diffusion for Policy Learning
Calvin Luo, Mandy He, Zilai Zeng, Chen Sun
Reinforcement Learning Text Guided Diffusion Text Based Reinforcement Learning Zero Shot Reward
Research on Autonomous Robots Navigation based on Reinforcement Learning
Zixiang Wang, Hao Yan, Yining Wang, Zhengjia Xu, Zhuoyue Wang, Zhizhong Wu
Reinforcement Learning Autonomous Navigation Proximal Policy Optimization DH Research Autonomous Robot Navigation

July 1, 2024

To Switch or Not to Switch? Balanced Policy Switching in Offline Reinforcement Learning
Tao Ma, Xuzhi Yang, Zoltan Szabo
Reinforcement Learning Offline Reinforcement Learning Optimal Transport Actor Critic Algorithm Optimal Behavior Policy Switching
Normalization and effective learning rates in reinforcement learning
Clare Lyle, Zeyu Zheng, Khimya Khetarpal, James Martens, Hado van Hasselt, Razvan Pascanu, Will Dabney
Reinforcement Learning Deep Reinforcement Learning Learning Rate Normalization Dictionary Normalization Layer Plasticity Loss
Weight Clipping for Deep Continual and Reinforcement Learning
Mohamed Elsayed, Qingfeng Lan, Clare Lyle, A. Rupam Mahmood
Reinforcement Learning Strong Generalization Continual Learning Neural Network Weight Weight Pruning
DogeRM: Equipping Reward Models with Domain Knowledge through Model Merging
Tzu-Han Lin, Chen-An Li, Hung-yi Lee, Yun-Nung Chen
Reinforcement Learning Domain Knowledge Reward Model Model Merging Specific Preference
Residual-MPPI: Online Policy Customization for Continuous Control
Pengcheng Wang, Chenran Li, Catherine Weaver, Kenta Kawamoto, Masayoshi Tomizuka, Chen Tang, Wei Zhan
Reinforcement Learning Zero Shot Imitation Learning Continuous Control Residual Stream Online Policy Customization Continuous Control Policy

June 30, 2024

Benchmarks for Reinforcement Learning with Biased Offline Data and Imperfect Simulators
Ori Linial, Guy Tennenholtz, Uri Shalit
Reinforcement Learning New Benchmark Offline Reinforcement Learning Model Based Offline Reinforcement Learning Offline Data Accurate Simulation
Model-Free Active Exploration in Reinforcement Learning
Alessio Russo, Alexandre Proutiere
Reinforcement Learning Ensemble Based Based Exploration Continuous Time Markov Decision Process
DEAR: Disentangled Environment and Agent Representations for Reinforcement Learning without Reconstruction
Ameya Pore, Riccardo Muradore, Diego Dall'Alba
Reinforcement Learning Full State Reconstruction Environment Feature Disentangled Representation Robotic Task Visual Reinforcement Learning Disentangled Haze Agent Representation
Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning
Yuheng Zhang, Dian Yu, Baolin Peng, Linfeng Song, Ye Tian, Mingyue Huo, Nan Jiang, Haitao Mi, Dong Yu
Large Language Model Reinforcement Learning Preference Feedback Game Theoretic Policy Iteration

June 28, 2024

External Model Motivated Agents: Reinforcement Learning for Enhanced Environment Sampling
Rishav Bhagat, Jonathan Balloch, Zhiyu Lin, Julia Kim, Mark Riedl
Reinforcement Learning Model Adaptation Model Based Agent Enhanced Sampling Influence Diagram Effective Adaptation Interest Modeling
PoliFormer: Scaling On-Policy RL with Transformers Results in Masterful Navigators
Kuo-Hao Zeng, Zichen Zhang, Kiana Ehsani, Rose Hendrix, Jordi Salvador, Alvaro Herrasti, Ross Girshick, Aniruddha Kembhavi, Luca Weihs
Reinforcement Learning Human Like RL Navigation Agent Transformer Based Policy Indoor Robot Navigation Multi Object Navigation
Applying RLAIF for Code Generation with API-usage in Lightweight LLMs
Sujan Dutta, Sayantan Mahinder, Raviteja Anantha, Bortik Bandyopadhyay
Reinforcement Learning Code Generation Code Generation Task API Usage AI Feedback Code Generation Ability Code Efficiency Lightweight LLM LLM Based Baseline
Operator World Models for Reinforcement Learning
Pietro Novelli, Marco Pratticò, Massimiliano Pontil, Carlo Ciliberto
Reinforcement Learning Sequential Decision Policy Mirror Descent Action Value Function
3D Operation of Autonomous Excavator based on Reinforcement Learning through Independent Reward for Individual Joints
Yoonkyu Yoo, Donghwi Jung, Seong-Woo Kim
Reinforcement Learning Learning Based Human Joint

Reinforcement Learning

Papers

Reinforcement Learning from Human Feedback: Whose Culture, Whose Values, Whose Perspectives?

Text-Aware Diffusion for Policy Learning

Research on Autonomous Robots Navigation based on Reinforcement Learning

To Switch or Not to Switch? Balanced Policy Switching in Offline Reinforcement Learning

Normalization and effective learning rates in reinforcement learning

Weight Clipping for Deep Continual and Reinforcement Learning

DogeRM: Equipping Reward Models with Domain Knowledge through Model Merging

Residual-MPPI: Online Policy Customization for Continuous Control

Benchmarks for Reinforcement Learning with Biased Offline Data and Imperfect Simulators

Model-Free Active Exploration in Reinforcement Learning

DEAR: Disentangled Environment and Agent Representations for Reinforcement Learning without Reconstruction

Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning

A Two-stage Reinforcement Learning-based Approach for Multi-entity Task Allocation

PUZZLES: A Benchmark for Neural Algorithmic Reasoning

Digital Twin-Assisted Data-Driven Optimization for Reliable Edge Caching in Wireless Networks

External Model Motivated Agents: Reinforcement Learning for Enhanced Environment Sampling

PoliFormer: Scaling On-Policy RL with Transformers Results in Masterful Navigators

Applying RLAIF for Code Generation with API-usage in Lightweight LLMs

Operator World Models for Reinforcement Learning

3D Operation of Autonomous Excavator based on Reinforcement Learning through Independent Reward for Individual Joints