Advantage Actor Critic

Advantage Actor-Critic (A2C) is a reinforcement learning algorithm aiming to improve the efficiency and stability of policy optimization by combining an actor network (for selecting actions) and a critic network (for evaluating actions). Current research focuses on enhancing A2C's performance through techniques like experience replay, variance reduction, and hybrid quantum-classical implementations, as well as extending its application to multi-agent systems and interpretability. These advancements are significant for improving the sample efficiency and applicability of reinforcement learning in diverse fields, including robotics, game playing, and recommender systems.

Papers

May 24, 2024

eQMARL: Entangled Quantum Multi-Agent Reinforcement Learning for Distributed Cooperation over Quantum Channels
Alexander DeRieux, Walid Saad
Multi Agent Reinforcement Learning Quantum Computing Quantum Physic Long Term Cooperation Quantum Reinforcement Learning Quantum Channel Advantage Actor Critic

February 15, 2024

Revisiting Experience Replayable Conditions
Taisuke Kobayashi
Reinforcement Learning Policy Reinforcement Learning Experience Replay Advantage Actor Critic

January 13, 2024

Quantum Advantage Actor-Critic for Reinforcement Learning
Michael Kölle, Mohamad Hgog, Fabian Ritz, Philipp Altmann, Maximilian Zorn, Jonas Stein, Claudia Linnhoff-Popien
Reinforcement Learning Variational Quantum Circuit Quantum Advantage Quantum Reinforcement Learning Advantage Actor Critic

September 9, 2023

Advantage Actor-Critic with Reasoner: Explaining the Agent's Behavior from an Exploratory Perspective
Muzhe Guo, Feixu Yu, Tian Lan, Fang Jin
Reinforcement Learning Agent Smith BEHAVIOR Explanation Exploratory Study Advantage Feedback Contrastive Reasoner Decision Problem Advantage Actor Critic

May 9, 2023

Latent Interactive A2C for Improved RL in Open Many-Agent Systems
Keyang He, Prashant Doshi, Bikramjit Banerjee
Latent Representation Decentralized Training Advantage Actor Critic Open Multi Agent System Apical 2 Chamber

November 29, 2022

Autotuning PID control using Actor-Critic Deep Reinforcement Learning
Vivien van Veldhuizen
Reinforcement Learning Actor Critic Adaptive Controller Advantage Actor Critic

November 19, 2022

Let Graph be the Go Board: Gradient-free Node Injection Attack for Graph Neural Networks via Reinforcement Learning
Mingxuan Ju, Yujie Fan, Chuxu Zhang, Yanfang Ye
Reinforcement Learning Graph Neural Network Graph Drawing Node Injection Go Game Structural Attack Advantage Actor Critic

June 14, 2022

Variance Reduction for Policy-Gradient Methods via Empirical Variance Minimization
Maxim Kaledin, Alexander Golubev, Denis Belomestny
Reinforcement Learning Policy Gradient Empirical Risk Minimization Variance Reduction Advantage Actor Critic EV GP Criterion

May 18, 2022

A2C is a special case of PPO
Shengyi Huang, Anssi Kanervisto, Antonin Raffin, Weixun Wang, Santiago Ontañón, Rousslan Fernand Julien Dossa
Proximal Policy Optimization Policy Optimization Game AI Deep Reinforcement Learning Algorithm Advantage Actor Critic Special Case Pseudo Code Apical 2 Chamber

April 19, 2022

Network Topology Optimization via Deep Reinforcement Learning
Zhuoran Li, Xing Wang, Ling Pan, Lin Zhu, Zhendong Wang, Junlan Feng, Chao Deng, Longbo Huang
Deep Reinforcement Learning Network Topology Arbitrary Topology Advantage Actor Critic Topology Optimisation

January 15, 2022

Recursive Least Squares Advantage Actor-Critic Algorithms
Yuan Wang, Chunyuan Zhang, Tianzong Yu, Meng Ma
Deep Reinforcement Learning Least Square Actor Critic Algorithm Recursive Algorithm Traditional Reinforcement Learning Advantage Actor Critic Apical 2 Chamber

November 5, 2021

Supervised Advantage Actor-Critic for Recommender Systems
Xin Xin, Alexandros Karatzoglou, Ioannis Arapakis, Joemon M. Jose
Reinforcement Learning Recommender System Sequential Recommendation Sequential Learning Advantage Actor Critic