Multiagent Bidirectionally-Coordinated Nets for Learning to Play StarCraft Combat Games

By Peng Peng, Quan Yuan, Ying Wen, Yaodong Yang, Zhenkun Tang, Haitao Long, Jun Wang. 2017

216 citations recorded by the source corpus.

How researchers see this paper

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

In Reinforcement Learning

Citations in the corpus

VAIN: Attentional Multi-agent Predictive Modeling (2017, neutral): “Other works dealing with multi-agent reinforcement learning include [22] and [23].”
Counterfactual Multi-Agent Reinforcement Learning with Graph Convolution Communication (2020, neutral): “CommNet [19] and BiCNet [17] communicate the encoding of local observation among all agents.”
Value Function Transfer for Deep Multi-Agent Reinforcement Learning Based on N-Step Returns (2019, background): “BiCNet [Peng et al. , 2017] uses bidirectional RNNs to communicate with others.”
Deep Multi-Agent Reinforcement Learning with Discrete-Continuous Hybrid Action Spaces (2019, background): “Reinforcement learning (RL) has recently shown a great success on a variety of cooperative multi-agent problems, such as multiplayer games [Peng et al. , 2017], autonomous cars [Cao et al. , 2013] and network packet delivery [Ye et al. , 2015].”
Multi-Agent Common Knowledge Reinforcement Learning (2018, background): “Peng et al. (2017) train a fully centralised controller that utilises a bidirectional RNN as a communication channel between agents, and decompose the global reward into per-agent local rewards.”
Mean Field Multi-Agent Reinforcement Learning (2018, background): “Researchers have also adopted the paradigm of centralized training with decentralized execution for multiagent policy-gradient learning: BICNET (Peng et al., 2017), COMA (Foerster et al.”
MAgent: A Many-Agent Reinforcement Learning Platform for Artificial Collective Intelligence (2017, background): “With certain rudimentary abilities, Artificial Collective Intelligence (ACI) starts to emerge from multiple domains, including stock trading (Wang and others 2013), strategic games (Peng et al. 2017) and city transportation optimization (Zhang et al.”
StarCraft II: A New Challenge for Reinforcement Learning (2017, background): “, [22, 36]) or learning game outcome or build orders from replay data [9, 12].”
Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environments (2017, background): “…(e.g., changes in the physics or characteristics of the agent), environments with multiple agents are particularly challenging due to complexity of the emergent behavior and are of practical interest with applications ranging from multiplayer games [16] to coordinating self-driving ﬂeets [17].”
A Brief Survey of Deep Reinforcement Learning (2017, neutral): “to agents sequentially [28], using a bidirectional channel (providing ordering with less signal loss) [87], and an all-toall channel [114].”
A survey and critique of multiagent deep reinforcement learning (2018, methodology): “ training a single network in which agents share their weights. Note that, since agents could receive dierent observations (e.g., in partially observable scenarios), they can still behave dierently [59, 60, 61, 62, 67, 68]. Recurrent networks. LSTM networks [39, 109] are widely used recurrent neural”
Learning to Schedule Communication in Multi-agent Reinforcement Learning (2019, background): “Foer-ster et al. (2016); Sukhbaatar et al. (2016); Peng et al. (2017); Guestrin et al. (2002), and Zhang & Lesser (2013) train multiple agents to learn a communication protocol, and have shown that communicating agents achieve better rewards at various tasks.”
When Does Communication Learning Need Hierarchical Multi-Agent Deep Reinforcement Learning (2019, methodology): “Still, the communication in CommNet and StarCraft is permanently available and the network decides on an action directly after communicating.”
An Empirical Study of AI Population Dynamics with Million-agent Reinforcement Learning (2017, background): “The work in (Peng et al. 2017) employed bidirectional recurrent neural networks to coordinate groups of agents to play StarCraft combat games, and achieved human-level micro-management skills.”
Hierarchical Decision Making by Generating and Following Natural Language Instructions (2019, background): “Inspired by the recent success of deep reinforcement learning, more works focus on training a neural network to finish sub-tasks [17, 24], some with strong computational requirement [28].”
Arena: A General Evaluation Platform and Building Toolkit for Multi-Agent Intelligence (2019, background): “However, collaborative behaviors are hardly observed under such fully decentralized setting; thus, a fully centralized system is utilized in [73, 96], where the policy has access to the global state and is shared by all agents.”