A Concise Introduction to Decentralized POMDPs

By Frans Adriaan Oliehoek, Chris Amato. 2016

1,383 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Management Science and Operations Research, Reinforcement Learning, Game & Decision Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

In Reinforcement Learning

In Game & Decision Theory

Citations in the corpus

The Surprising Effectiveness of PPO in Cooperative Multi-Agent Games (2021, methodology): “We study decentralized partially observable Markov decision processes (DEC-POMDP) [24] with shared rewards.”
Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning (2020, background): “…cooperative multi-agent sequential decision-making task can be described as a decentralised partially observable Markov decision process (Dec-POMDP) (Oliehoek and Amato, 2016) consisting of a tuple G = (cid:104) S, U, P, r, Z, O, n, γ (cid:105) . s ∈ S describes the true state of the environment.”
Value-Decomposition Networks For Cooperative Multi-Agent Learning (2017, background): “This is consistent with the Dec-POMDP framework (Oliehoek et al., 2008; Oliehoek and Amato, 2016).”
Deep Decentralized Multi-task Multi-Agent Reinforcement Learning under Partial Observability (2017, background): “Efﬁcient learners must extract knowledge from past tasks to accelerate learning and improve generalization to new tasks.”
TiKick: Towards Playing Multi-agent Football Full Games from Single-agent Demonstrations (2021, background): “The multi-agent reinforcement learning can be formalized as Dec-POMDPs (Oliehoek and Amato, 2016).”
Celebrating Diversity in Shared Multi-Agent Reinforcement Learning (2021, background): “A fully cooperative multi-agent task can be formulated as a Dec-POMDP [14], which is defined as a tuple G = 〈N,S,A, P,R,O,Ω, n, γ〉, where N is a finite set of n agents, s ∈ S is the true state of the environment, A is the set of actions, and γ ∈ [0, 1) is a discount factor.”
The StarCraft Multi-Agent Challenge (2019, methodology): “We brieﬂy review the formalism of such fully cooperative multi-agent tasks as Dec-POMDPs but refer readers to Oliehoek & Amato (2016) for a more complete picture.”
QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning (2018, background): “A fully cooperative multi-agent task can be described as a Dec-POMDP (Oliehoek & Amato, 2016) Although training is centralised, execution is decentralised, i.e., the learning algorithm has access to all local action-observation histories τ and global state s , but each agent’s learnt policy can…”
Deep Coordination Graphs (2019, methodology): “In this paper we assume a Dec-POMDP for n agents (cid:104)S , {A i } ni =1 , P, r, & Amato, 2016).”
Decentralized Likelihood Quantile Networks for Improving Performance in Deep Multi-Agent Reinforcement Learning (2018, background): “Furthermore, Dec-POMDPs generalize POMDPs to decentralized settings (Oliehoek & Amato, 2016) with multiple cooperative agents.”
Machine Theory of Mind (2018, background): “Learning rich models of others will improve decision-making in complex multi-agent tasks, especially where model-based planning and imagination are required (Hassabis et al., 2013; Hula et al., 2015; Oliehoek & Amato, 2016).”
A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning (2017, background): “There is a significant amount of work that deals with the case of decentralized cooperative problems [76, 79], and in the general setting by extending the notion of belief states and Bayesian updating from POMDPs [28].”