Multi-agent deep reinforcement learning: a survey

By Sven Gronauer, Klaus-Jürgen Diepold. Artificial Intelligence Review, 2021

931 citations recorded by the source corpus.

The advances in reinforcement learning have recorded sublime success in various domains. Although the multi-agent domain has been overshadowed by its single-agent counterpart during this progress, multi-agent reinforcement learning gains rapid traction, and the latest accomplishments address problems with real-world complexity. This article provides an overview of the current developments in the field of multi-agent deep reinforcement learning. We focus primarily on literature from recent years that combines deep reinforcement learning methods with a multi-agent scenario. To survey the works that constitute the contemporary landscape, the main contents are divided into three parts. First, we analyze the structure of training schemes that are applied to train multiple agents. Second, we consider the emergent patterns of agent behavior in cooperative, competitive and mixed scenarios. Third, we systematically enumerate challenges that exclusively arise in the multi-agent domain and review methods that are leveraged to cope with these challenges. To conclude this survey, we discuss advances, identify trends, and outline possible directions for future work in this research area.

Subfields: Artificial Intelligence, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/98cb29c03a0882fde368db28ade214c57c3239d6

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 Control and Systems Engineering

Citations in the corpus

Explainable artificial intelligence: a comprehensive review (2021, background): “It is recommended that the interested readers referred to the latest reviews of reinforcement learning (Gronauer and Diepold 2021) and meta-learning (Huisman et al.”
A survey of multi-agent deep reinforcement learning with communication (2022, neutral): “Based on the three key components, we identify and summarize 9 research questions that commonly arise in Comm-MADRL system design, corresponding to 9 dimensions as detailed in Table 1.”
Human-Timescale Adaptation in an Open-Ended Task Space (2023, background): “Division of labour and physical coordination have received extensive study in the multi-agent RL literature, but prior approaches have tended to focus on speciﬁc domains rather than training generalists, e.g. (Wang et al., 2020b; Yang et al., 2020; Strouse et al., 2021; Gronauer and Diepold, 2022).”