Double Deep Q Learning

Double Deep Q-learning (DDQL) is a reinforcement learning technique aiming to improve the stability and efficiency of deep Q-networks by mitigating overestimation bias inherent in standard Q-learning. Current research focuses on enhancing DDQL's performance in various applications, including continuous control tasks, multi-agent systems, and federated learning, often incorporating modifications like prioritized experience replay, actor-critic methods, and novel loss functions to accelerate learning and improve robustness. This approach shows promise across diverse fields, from optimizing energy production in wind turbines and improving water quality monitoring to accelerating drug discovery and enhancing resource allocation in next-generation communication networks.

Papers

October 15, 2024

Improve Value Estimation of Q Function and Reshape Reward with Monte Carlo Tree Search
Jiamian Li
Reinforcement Learning Monte Carlo Tree Search Reward Shaping Q Function Imperfect Information Game Double Deep Q Learning Value Estimation Fictitious Self Play

October 11, 2024

Hybrid LLM-DDQN based Joint Optimization of V2I Communication and Autonomous Driving
Zijiang Yan, Hao Zhou, Hina Tabassum, Xue Liu
Autonomous Driving Joint Optimization Vehicular Network Vehicle 2 Vehicle Double Deep Q Learning

February 17, 2024

Reinforcement learning to maximise wind turbine energy generation
Daniel Soler, Oscar Mariño, David Huergo, Martín de Frutos, Esteban Ferrer
Reinforcement Learning Reinforcement Learning Approach Wind Turbine Double Deep Q Learning

February 14, 2024

Exploiting Estimation Bias in Deep Double Q-Learning for Actor-Critic Methods
Alberto Sinigaglia, Niccolò Turcato, Alberto Dalla Libera, Ruggero Carli, Gian Antonio Susto
Reinforcement Learning Policy Learning Actor Critic Estimation Bias Double Deep Q Learning Delayed Deep Deterministic Policy Gradient

January 9, 2024

Deep Reinforcement Multi-agent Learning framework for Information Gathering with Local Gaussian Processes for Water Monitoring
Samuel Yanes Luis, Dmitriy Shutin, Juan Marchal Gómez, Daniel Gutiérrez Reina, Sergio Toral Marín
Deep Reinforcement Learning Gaussian Process Multi Agent Deep Reinforcement Learning Autonomous Surface Vehicle Water Quality Information Acquisition Double Deep Q Learning

November 12, 2023

FLASH-RL: Federated Learning Addressing System and Static Heterogeneity using Reinforcement Learning
Sofiane Bouaziz, Hadjer Benmeziane, Youcef Imine, Leila Hamdad, Smail Niar, Hamza Ouarnoughi
Reinforcement Learning Mobile User Unknown Heterogeneity Double Deep Q Learning

October 30, 2023

DGFN: Double Generative Flow Networks
Elaine Lau, Nikhil Vemgal, Doina Precup, Emmanuel Bengio
Drug Discovery Generative Flow Network Energy Based GFlowNets Gabor Filter Double Deep Q Learning

September 18, 2023

August 28, 2023

Reinforcement Learning for Sampling on Temporal Medical Imaging Sequences
Zhishen Huang
Reinforcement Learning Open Sampling Reconstruction Algorithm Optimal Sampling Double Deep Q Learning

September 16, 2022

M$^2$DQN: A Robust Method for Accelerating Deep Q-learning Network
Zhe Zhang, Yukun Zou, Junjie Lai, Qing Xu
Deep Q Network Deep Q Learning DQN Agent Robust Method Double Deep Q Learning

May 24, 2022

Deep Reinforcement Learning for Multi-class Imbalanced Training
Jenny Yang, Rasheed El-Bouri, Odhran O'Donoghue, Alexander S. Lachapelle, Andrew A. S. Soltan, David A. Clifton
Reinforcement Learning Deep Reinforcement Learning Imbalanced Learning Double Deep Q Learning Class Imbalanced Learning Balanced Classification

January 21, 2022

Deep Q-learning: a robust control approach
Balazs Varga, Balazs Kulcsar, Morteza Haghir Chehreghani
Reinforcement Learning Learning Dynamic Deep Q Learning Robust Control Robust Controller Double Deep Q Learning

Double Deep Q Learning

Papers

Improve Value Estimation of Q Function and Reshape Reward with Monte Carlo Tree Search

Hybrid LLM-DDQN based Joint Optimization of V2I Communication and Autonomous Driving

Reinforcement learning to maximise wind turbine energy generation

Exploiting Estimation Bias in Deep Double Q-Learning for Actor-Critic Methods

Deep Reinforcement Multi-agent Learning framework for Information Gathering with Local Gaussian Processes for Water Monitoring

FLASH-RL: Federated Learning Addressing System and Static Heterogeneity using Reinforcement Learning

DGFN: Double Generative Flow Networks

Double Deep Q-Learning-based Path Selection and Service Placement for Latency-Sensitive Beyond 5G Applications

Self-Sustaining Multiple Access with Continual Deep Reinforcement Learning for Dynamic Metaverse Applications

Reinforcement Learning for Sampling on Temporal Medical Imaging Sequences

M$^2$DQN: A Robust Method for Accelerating Deep Q-learning Network

Deep Reinforcement Learning for Multi-class Imbalanced Training

Deep Q-learning: a robust control approach