Open AccessProceedings Article
QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning
Tabish Rashid,Mikayel Samvelyan,Christian Schroeder,Gregory Farquhar,Jakob Foerster,Shimon Whiteson +5 more
- pp 4292-4301
TLDR
QMIX employs a network that estimates joint action-values as a complex non-linear combination of per-agent values that condition only on local observations, and structurally enforce that the joint-action value is monotonic in the per- agent values, which allows tractable maximisation of the jointaction-value in off-policy learning.About:
This article is published in International Conference on Machine Learning.The article was published on 2018-07-03 and is currently open access. It has received 505 citations till now. The article focuses on the topics: Reinforcement learning & Monotonic function.read more
Citations
More filters
Posted Content
A review of mobile robot motion planning methods: from classical motion planning workflows to reinforcement learning-based architectures.
TL;DR: In this paper, the authors provide a systematic review of various RL-based motion planning methods, including RL optimization motion planners, map-free end-to-end methods that integrate sensing and decision-making and multi-robot cooperative planning methods.
Posted Content
Noisy-MAPPO: Noisy Credit Assignment for Cooperative Multi-agent Actor-Critic methods
Siyue Hu,Jian Hu,Shih-Wei Liao +2 more
TL;DR: In this article, the authors theoretically generalize PPO to MAPPO by a approximate lower bound of Trust Region Policy Optimization (TRPO), which guarantees its convergence, and propose the noisy credit assignment methods (NoisyMAPPO and Advantage-Noisy-MAPPO) to solve it.
Posted Content
Informative Policy Representations in Multi-Agent Reinforcement Learning via Joint-Action Distributions.
TL;DR: In this article, the authors propose a general method to learn representations of other agents' policies via the joint-action distributions sampled in interactions, which can well generalize to unseen agents.
Posted Content
Reinforcement Learning In Two Player Zero Sum Simultaneous Action Games.
TL;DR: In this paper, the authors introduce two RL agents, Meta-Nash DQN and Best Response AgenT (BRAT), to find the best response to exploit the opponent's strategy.
Posted Content
Regularize! Don't Mix: Multi-Agent Reinforcement Learning without Explicit Centralized Structures.
TL;DR: In this article, the authors propose using regularization for multi-agent reinforcement learning rather than learning explicit cooperative structures called Multi-Agent Regularized Q-learning (MARQ), which leverages shared experiences of the agents to regularize the individual policies in order to promote structured exploration.
References
More filters
Journal ArticleDOI
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Journal ArticleDOI
Human-level control through deep reinforcement learning
Volodymyr Mnih,Koray Kavukcuoglu,David Silver,Andrei Rusu,Joel Veness,Marc G. Bellemare,Alex Graves,Martin Riedmiller,Andreas K. Fidjeland,Georg Ostrovski,Stig Petersen,Charles Beattie,Amir Sadik,Ioannis Antonoglou,Helen King,Dharshan Kumaran,Daan Wierstra,Shane Legg,Demis Hassabis +18 more
TL;DR: This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.
Posted Content
Empirical evaluation of gated recurrent neural networks on sequence modeling
TL;DR: These advanced recurrent units that implement a gating mechanism, such as a long short-term memory (LSTM) unit and a recently proposed gated recurrent unit (GRU), are found to be comparable to LSTM.
Deep reinforcement learning with double Q-learning
TL;DR: In this article, the authors show that the DQN algorithm suffers from substantial overestimation in some games in the Atari 2600 domain, and they propose a specific adaptation to the algorithm and show that this algorithm not only reduces the observed overestimations, but also leads to much better performance on several games.
Journal ArticleDOI
Learning from delayed rewards
TL;DR: The invention relates to a circuit for use in a receiver which can receive two-tone/stereo signals which is intended to make a choice between mono or stereo reproduction of signal A or of signal B and vice versa.