Curriculum Reinforcement Learning

Curriculum reinforcement learning (CRL) aims to accelerate the training of reinforcement learning agents by structuring the learning process, progressing from simpler to more complex tasks. Current research focuses on automating curriculum generation, often employing techniques like optimal transport, goal-conditional policies, and various model architectures (e.g., mixture of experts, quantized world models) to efficiently explore the task space and improve sample efficiency. This approach holds significant promise for tackling complex real-world problems, particularly in robotics and autonomous systems, where data acquisition is expensive and traditional reinforcement learning methods often struggle with sample complexity. The development of robust and adaptable CRL methods is driving progress in diverse fields, including autonomous driving, robotic manipulation, and quantum computing.

Papers

June 9, 2023

Self-Paced Absolute Learning Progress as a Regularized Approach to Curriculum Learning
Tobias Niehues, Ulla Scheler, Pascal Klink
Reinforcement Learning Neural Network LeArning Abstract Curriculum Learning Curriculum Reinforcement Learning Learning Progress

March 9, 2023

GOATS: Goal Sampling Adaptation for Scooping with Curriculum Reinforcement Learning
Yaru Niu, Shiyu Jin, Zeqing Zhang, Jiacheng Zhu, Ding Zhao, Liangjun Zhang
Goal Conditioned Reinforcement Learning State Action Space Real World Scenario Curriculum Reinforcement Learning

March 7, 2023

Chance-Aware Lane Change with High-Level Model Predictive Control Through Curriculum Reinforcement Learning
Yubin Wang, Yulin Li, Zengqi Peng, Hakim Ghazzai, Jun Ma
Episodic Reinforcement Learning Curriculum Reinforcement Learning

February 25, 2023

Autonomous Exploration and Mapping for Mobile Robots via Cumulative Curriculum Reinforcement Learning
Zhi Li, Jinghao Xin, Ning Li
Deep Reinforcement Learning Mobile Robot Area MaPPing Autonomous Exploration State Representation Curriculum Reinforcement Learning

December 29, 2022

Backward Curriculum Reinforcement Learning
KyungMin Ko
Reinforcement Learning Reward Shaping Curriculum Reinforcement Learning

October 18, 2022

Curriculum Reinforcement Learning using Optimal Transport via Gradual Domain Adaptation
Peide Huang, Mengdi Xu, Jiacheng Zhu, Laixi Shi, Fei Fang, Ding Zhao
Optimal Transport Natural Gradient Generalized Geodesic Curriculum Reinforcement Learning Task Distribution Gradual Domain Adaptation

August 4, 2022

Human Decision Makings on Curriculum Reinforcement Learning with Difficulty Adjustment
Yilei Zeng, Jiali Duan, Yang Li, Emilio Ferrara, Lerrel Pinto, C. -C. Jay Kuo, Stefanos Nikolaidis
Reinforcement Learning Human Decision Human Centered Curriculum Reinforcement Learning Large Scale Reinforcement Learning Reinforcement Learning Performance Difficulty Adjustment

July 10, 2022

State Dropout-Based Curriculum Reinforcement Learning for Self-Driving at Unsignalized Intersections
Shivesh Khaitan, John M. Dolan
Deep Reinforcement Learning Autonomous Driving Autonomous Vehicle Internal State Curriculum Reinforcement Learning Unsignalized Intersection Autonomous Driving Task

June 10, 2022

ROI-Constrained Bidding via Curriculum-Guided Bayesian Reinforcement Learning
Haozhe Wang, Chao Du, Panyan Fang, Shuo Yuan, Xuming He, Liang Wang, Bo Zheng
Observable Markov Decision Process Curriculum Reinforcement Learning Paper Bidding Cost Constraint Real Time Bidding Ad Market

May 20, 2022

Learning Progress Driven Multi-Agent Curriculum
Wenshuai Zhao, Zhiyuan Li, Joni Pajarinen
Multi Agent Reinforcement Learning Single Agent Curriculum Reinforcement Learning Training Agent Strong Learner

April 26, 2022

Collaborative Target Search with a Visual Drone Swarm: An Adaptive Curriculum Embedded Multistage Reinforcement Learning Approach
Jiaping Xiao, Phumrapee Pisutsin, Mir Feroskhan
Curriculum Reinforcement Learning Multi Drone Drone Swarm Target Search

March 29, 2022

Assessing Evolutionary Terrain Generation Methods for Curriculum Reinforcement Learning
David Howard, Josh Kannemeyer, Davide Dolcetti, Humphrey Munn, Nicole Robinson
GAN Model Curriculum Learning Curriculum Reinforcement Learning Terrain Generation

December 18, 2021

Curriculum Based Reinforcement Learning of Grid Topology Controllers to Prevent Thermal Cascading
Amarsagar Reddy Ramapuram Matavalam, Kishan Prudhvi Guddanti, Yang Weng, Venkataramana Ajjarapu
Reinforcement Learning Cascade Model Curriculum Reinforcement Learning Real World Reinforcement Learning Training Agent Power Network

November 19, 2021

Reinforcement Learning with Adaptive Curriculum Dynamics Randomization for Fault-Tolerant Robot Control
Wataru Okamoto, Hiroshi Kera, Kazuhiko Kawamoto
Reinforcement Learning Robotic System Curriculum Reinforcement Learning Fault Tolerant Control Lloyd Algorithm Aware Curriculum Actuator Fault

Curriculum Reinforcement Learning

Papers

Self-Paced Absolute Learning Progress as a Regularized Approach to Curriculum Learning

GOATS: Goal Sampling Adaptation for Scooping with Curriculum Reinforcement Learning

Chance-Aware Lane Change with High-Level Model Predictive Control Through Curriculum Reinforcement Learning

Autonomous Exploration and Mapping for Mobile Robots via Cumulative Curriculum Reinforcement Learning

Backward Curriculum Reinforcement Learning

Curriculum Reinforcement Learning using Optimal Transport via Gradual Domain Adaptation

Human Decision Makings on Curriculum Reinforcement Learning with Difficulty Adjustment

State Dropout-Based Curriculum Reinforcement Learning for Self-Driving at Unsignalized Intersections

ROI-Constrained Bidding via Curriculum-Guided Bayesian Reinforcement Learning

Learning Progress Driven Multi-Agent Curriculum

Collaborative Target Search with a Visual Drone Swarm: An Adaptive Curriculum Embedded Multistage Reinforcement Learning Approach

Assessing Evolutionary Terrain Generation Methods for Curriculum Reinforcement Learning

Curriculum Based Reinforcement Learning of Grid Topology Controllers to Prevent Thermal Cascading

Reinforcement Learning with Adaptive Curriculum Dynamics Randomization for Fault-Tolerant Robot Control