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

October 4, 2024

Spatial-aware decision-making with ring attractors in reinforcement learning systems
Marcos Negre Saura, Richard Allmendinger, Theodore Papamarkou, Wei Pan
Reinforcement Learning Deep Reinforcement Learning Action Selection Reinforcement Learning System Global Attractor Neuronal Dynamic
Hybrid Classical/RL Local Planner for Ground Robot Navigation
Vishnu D. Sharma, Jeongran Lee, Matthew Andrews, Ilija Hadžić
Reinforcement Learning Bridging Text Dynamic Obstacle Path Tracking Hybrid Planning Autonomous Ground Navigation

October 3, 2024

Finite-Sample Analysis of the Monte Carlo Exploring Starts Algorithm for Reinforcement Learning
Suei-Wen Chen, Keith Ross, Pierre Youssef
Reinforcement Learning Monte Carlo Policy Iteration Finite Sample
Solving Reach-Avoid-Stay Problems Using Deep Deterministic Policy Gradients
Gabriel Chenevert, Jingqi Li, Achyuta kannan, Sangjae Bae, Donggun Lee
Reinforcement Learning High Dimensional Reachability Analysis Deep Deterministic Policy Gradient Reach Avoid Invariant Set
ReLIC: A Recipe for 64k Steps of In-Context Reinforcement Learning for Embodied AI
Ahmad Elawady, Gunjan Chhablani, Ram Ramrakhya, Karmesh Yadav, Dhruv Batra, Zsolt Kira, Andrew Szot
Reinforcement Learning Embodied AI Complete Recipe Embodied Agent Ring Artifact Large Scale Reinforcement Learning Step Recognition
MA-RLHF: Reinforcement Learning from Human Feedback with Macro Actions
Yekun Chai, Haoran Sun, Huang Fang, Shuohuan Wang, Yu Sun, Hua Wu
Large Language Model Reinforcement Learning Human Feedback Reinforcement Learning From Human Feedback
Choices are More Important than Efforts: LLM Enables Efficient Multi-Agent Exploration
Yun Qu, Boyuan Wang, Yuhang Jiang, Jianzhun Shao, Yixiu Mao, Cheems Wang, Chang Liu, Xiangyang Ji
Large Language Model Reinforcement Learning Action Space Value Laden Choice Manual Effort Multi Agent Exploration
Dual Active Learning for Reinforcement Learning from Human Feedback
Pangpang Liu, Chengchun Shi, Will Wei Sun
Large Language Model Reinforcement Learning Offline Reinforcement Learning Human Feedback Reward Report Dual Learning Reward Estimator
Cross-Embodiment Dexterous Grasping with Reinforcement Learning
Haoqi Yuan, Bohan Zhou, Yuhui Fu, Zongqing Lu
Reinforcement Learning Robotic Hand Functional Grasping Robot Hand Dexterous Hand
End-to-end Driving in High-Interaction Traffic Scenarios with Reinforcement Learning
Yueyuan Li, Mingyang Jiang, Songan Zhang, Wei Yuan, Chunxiang Wang, Ming Yang
Reinforcement Learning Autonomous Driving End to End Traffic Scenario Large Scale Reinforcement Learning Optimal Driving
Doubly Optimal Policy Evaluation for Reinforcement Learning
Shuze Liu, Claire Chen, Shangtong Zhang
Reinforcement Learning Optimal Policy Policy Evaluation Data Processing
Abstract Reward Processes: Leveraging State Abstraction for Consistent Off-Policy Evaluation
Shreyas Chaudhari, Ameet Deshpande, Bruno Castro da Silva, Philip S. Thomas
Reinforcement Learning Policy Evaluation State Abstraction Geometric Median Policy Data

October 2, 2024

RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning
Jonas Gehring, Kunhao Zheng, Jade Copet, Vegard Mella, Taco Cohen, Gabriel Synnaeve
Large Language Model Reinforcement Learning End to End Code Synthesis Execution Based Competitive Programming
Don't flatten, tokenize! Unlocking the key to SoftMoE's efficacy in deep RL
Ghada Sokar, Johan Obando-Ceron, Aaron Courville, Hugo Larochelle, Pablo Samuel Castro
Reinforcement Learning Deep Neural Network Deep Reinforcement Learning Mixture Component Multiple Expert
LLM-Augmented Symbolic Reinforcement Learning with Landmark-Based Task Decomposition
Alireza Kheirandish, Duo Xu, Faramarz Fekri
Reinforcement Learning First Order Inductive Logic Programming Task Decomposition Logical Rule
PreND: Enhancing Intrinsic Motivation in Reinforcement Learning through Pre-trained Network Distillation
Mohammadamin Davoodabadi, Negin Hashemi Dijujin, Mahdieh Soleymani Baghshah
Reinforcement Learning Intrinsic Reward Intrinsic Motivation Random Network Distillation
Mimicking Human Intuition: Cognitive Belief-Driven Q-Learning
Xingrui Gu, Guanren Qiao, Chuyi Jiang, Tianqing Xia, Hangyu Mao
Reinforcement Learning Q Learning Human Intuition Belief Distribution
MOREL: Enhancing Adversarial Robustness through Multi-Objective Representation Learning
Sedjro Salomon Hotegni, Sebastian Peitz
Reinforcement Learning Adversarial Attack Adversarial Example Adversarial Robustness Adversarial Perturbation Multi Objective Adversarial Feature
Stable Offline Value Function Learning with Bisimulation-based Representations
Brahma S. Pavse, Yudong Chen, Qiaomin Xie, Josiah P. Hanna
Reinforcement Learning Value Function State Action Pair Bisimulation Metric Continual Offline Reinforcement Learning
From Reward Shaping to Q-Shaping: Achieving Unbiased Learning with LLM-Guided Knowledge
Xiefeng Wu
Large Language Model Reinforcement Learning Reward Shaping Unbiased Learning Q Value

Reinforcement Learning

Papers

Spatial-aware decision-making with ring attractors in reinforcement learning systems

Hybrid Classical/RL Local Planner for Ground Robot Navigation

Finite-Sample Analysis of the Monte Carlo Exploring Starts Algorithm for Reinforcement Learning

Solving Reach-Avoid-Stay Problems Using Deep Deterministic Policy Gradients

ReLIC: A Recipe for 64k Steps of In-Context Reinforcement Learning for Embodied AI

MA-RLHF: Reinforcement Learning from Human Feedback with Macro Actions

Choices are More Important than Efforts: LLM Enables Efficient Multi-Agent Exploration

Dual Active Learning for Reinforcement Learning from Human Feedback

Cross-Embodiment Dexterous Grasping with Reinforcement Learning

End-to-end Driving in High-Interaction Traffic Scenarios with Reinforcement Learning

Doubly Optimal Policy Evaluation for Reinforcement Learning

Abstract Reward Processes: Leveraging State Abstraction for Consistent Off-Policy Evaluation

RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning

Don't flatten, tokenize! Unlocking the key to SoftMoE's efficacy in deep RL

LLM-Augmented Symbolic Reinforcement Learning with Landmark-Based Task Decomposition

PreND: Enhancing Intrinsic Motivation in Reinforcement Learning through Pre-trained Network Distillation

Mimicking Human Intuition: Cognitive Belief-Driven Q-Learning

MOREL: Enhancing Adversarial Robustness through Multi-Objective Representation Learning

Stable Offline Value Function Learning with Bisimulation-based Representations

From Reward Shaping to Q-Shaping: Achieving Unbiased Learning with LLM-Guided Knowledge