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 2, 2024

Adaptive teachers for amortized samplers
Minsu Kim, Sanghyeok Choi, Taeyoung Yun, Emmanuel Bengio, Leo Feng, Jarrid Rector-Brooks, Sungsoo Ahn, Jinkyoo Park, Nikolay Malkin, Yoshua Bengio
Reinforcement Learning Efficient Exploration Fast Sampling Adaptive Teaching
Scalable Reinforcement Learning-based Neural Architecture Search
Amber Cassimon, Siegfried Mercelis, Kevin Mets
Reinforcement Learning Neural Architecture Search Reinforcement Learning Agent Optimal Architecture
Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks
Xingxuan Li, Weiwen Xu, Ruochen Zhao, Fangkai Jiao, Shafiq Joty, Lidong Bing
Large Language Model Reinforcement Learning Complex Reasoning Task Planning Retrieval Augmentation Complex Task Critic Model
Sampling from Energy-based Policies using Diffusion
Vineet Jain, Tara Akhound-Sadegh, Siamak Ravanbakhsh
Reinforcement Learning Diffusion Explainer Diffusion Sampling Maximum Entropy Reinforcement Learning
Sparse Autoencoders Reveal Temporal Difference Learning in Large Language Models
Can Demircan, Tankred Saanum, Akshay K. Jagadish, Marcel Binz, Eric Schulz
Large Language Model Reinforcement Learning Context Learning Sparse Autoencoders
Real-World Data and Calibrated Simulation Suite for Offline Training of Reinforcement Learning Agents to Optimize Energy and Emission in Buildings for Environmental Sustainability
Judah Goldfeder, John Sipple
Reinforcement Learning Reinforcement Learning Agent Energy Policy Research Real World Data Model Calibration Smart Building Heating, Ventilation, and Air Conditioning Offline Training

October 1, 2024

Collaborative motion planning for multi-manipulator systems through Reinforcement Learning and Dynamic Movement Primitives
Siddharth Singh, Tian Xu, Qing Chang
Reinforcement Learning Motion Planning Collision Free Dynamic Movement Primitive Multiple Mobile Manipulator
Contrastive Abstraction for Reinforcement Learning
Vihang Patil, Markus Hofmarcher, Elisabeth Rumetshofer, Sepp Hochreiter
Reinforcement Learning Contrastive Learning Cognitive Abstraction State Representation Contrastive Training Abstraction Learning
A transformer-based deep reinforcement learning approach to spatial navigation in a partially observable Morris Water Maze
Marte Eggen, Inga Strümke
Reinforcement Learning Deep Reinforcement Learning Constructive Approach Transformer Architecture Local Navigation Maze Navigation Encoder Only Transformer Navigation Strategy
Enhancing Solution Efficiency in Reinforcement Learning: Leveraging Sub-GFlowNet and Entropy Integration
Siyi He
Reinforcement Learning Markov Chain Monte Carlo Molecule Generation Traditional Reinforcement Learning Enhancing Efficiency Trajectory Wise Reward

September 30, 2024

Demonstrating the Continual Learning Capabilities and Practical Application of Discrete-Time Active Inference
Rithvik Prakki
Reinforcement Learning Continual Learning Active Inference Textual Demonstration Practical Application Continual Adaptation Adaptive AI
Learning to Swim: Reinforcement Learning for 6-DOF Control of Thruster-driven Autonomous Underwater Vehicles
Levi Cai, Kevin Chang, Yogesh Girdhar
Reinforcement Learning LeArning Abstract Autonomous Underwater Vehicle 6 DoF
From homeostasis to resource sharing: Biologically and economically compatible multi-objective multi-agent AI safety benchmarks
Roland Pihlakas, Joel Pyykkö
Reinforcement Learning New Resource Multi Agent Collaboration AI Safety Human Value Safe AI Safety Benchmark Homeostatic Design Confers
Opt2Skill: Imitating Dynamically-feasible Whole-Body Trajectories for Versatile Humanoid Loco-Manipulation
Fukang Liu, Zhaoyuan Gu, Yilin Cai, Ziyi Zhou, Shijie Zhao, Hyunyoung Jung, Sehoon Ha, Yue Chen, Danfei Xu, Ye Zhao
Reinforcement Learning Loco Manipulation Whole Body Trajectory
The Perfect Blend: Redefining RLHF with Mixture of Judges
Tengyu Xu, Eryk Helenowski, Karthik Abinav Sankararaman, Di Jin, Kaiyan Peng, Eric Han, Shaoliang Nie, Chen Zhu, Hejia Zhang, Wenxuan Zhou, Zhouhao Zeng, Yun He, Karishma Mandyam, Arya Talabzadeh, Madian Khabsa, Gabriel Cohen, Yuandong Tian, Hao Ma, Sinong Wang, Han Fang
Reinforcement Learning Multi Task Learning Mixture Component Training Dynamic Reinforcement Learning From Human Feedback Expert Tax Judge
ACE: Abstractions for Communicating Efficiently
Jonathan D. Thomas, Andrea Silvi, Devdatt Dubhashi, Vikas Garg, Moa Johansson
Reinforcement Learning Timely Communication Conversational Agent Communication Efficient Metric Aware Abstraction Task Oriented Communication Semantic Abstraction

September 29, 2024

Adaptive Event-triggered Reinforcement Learning Control for Complex Nonlinear Systems
Umer Siddique, Abhinav Sinha, Yongcan Cao
Reinforcement Learning Control Policy Complex Dynamical System Natural Trigger Non Stationary Reinforcement Learning Continuous Time Dynamical System
LiRA: Light-Robust Adversary for Model-based Reinforcement Learning in Real World
Taisuke Kobayashi
Reinforcement Learning Real World Adversarial Learning Model Based Reinforcement Learning Quadrupedal Robot Dynamic Gait

September 28, 2024

Double Actor-Critic with TD Error-Driven Regularization in Reinforcement Learning
Haohui Chen, Zhiyong Chen, Aoxiang Liu, Wentuo Fang
Reinforcement Learning Actor Critic Framework Critic Regularization
Learning to Bridge the Gap: Efficient Novelty Recovery with Planning and Reinforcement Learning
Alicia Li, Nishanth Kumar, Tomás Lozano-Pérez, Leslie Kaelbling
Reinforcement Learning LeArning Abstract Task Planning Autonomous Robot Novelty Detection Model Based Planning Long Horizon Decision Making

Reinforcement Learning

Papers

Adaptive teachers for amortized samplers

Scalable Reinforcement Learning-based Neural Architecture Search

Can We Further Elicit Reasoning in LLMs? Critic-Guided Planning with Retrieval-Augmentation for Solving Challenging Tasks

Sampling from Energy-based Policies using Diffusion

Sparse Autoencoders Reveal Temporal Difference Learning in Large Language Models

Real-World Data and Calibrated Simulation Suite for Offline Training of Reinforcement Learning Agents to Optimize Energy and Emission in Buildings for Environmental Sustainability

Collaborative motion planning for multi-manipulator systems through Reinforcement Learning and Dynamic Movement Primitives

Contrastive Abstraction for Reinforcement Learning

A transformer-based deep reinforcement learning approach to spatial navigation in a partially observable Morris Water Maze

Enhancing Solution Efficiency in Reinforcement Learning: Leveraging Sub-GFlowNet and Entropy Integration

Demonstrating the Continual Learning Capabilities and Practical Application of Discrete-Time Active Inference

Learning to Swim: Reinforcement Learning for 6-DOF Control of Thruster-driven Autonomous Underwater Vehicles

From homeostasis to resource sharing: Biologically and economically compatible multi-objective multi-agent AI safety benchmarks

Opt2Skill: Imitating Dynamically-feasible Whole-Body Trajectories for Versatile Humanoid Loco-Manipulation

The Perfect Blend: Redefining RLHF with Mixture of Judges

ACE: Abstractions for Communicating Efficiently

Adaptive Event-triggered Reinforcement Learning Control for Complex Nonlinear Systems

LiRA: Light-Robust Adversary for Model-based Reinforcement Learning in Real World

Double Actor-Critic with TD Error-Driven Regularization in Reinforcement Learning

Learning to Bridge the Gap: Efficient Novelty Recovery with Planning and Reinforcement Learning