Robust Multi-agent Communication via Multi-view Message Certification
Not on the map. Why?
Abstract (arXiv)
Many multi-agent scenarios require message sharing among agents to promote coordination, hastening the robustness of multi-agent communication when policies are deployed in a message perturbation environment. Major relevant works tackle this issue under specific assumptions, like a limited number of message channels would sustain perturbations, limiting the efficiency in complex scenarios. In this paper, we take a further step addressing this issue by learning a robust multi-agent communication policy via multi-view message certification, dubbed CroMAC. Agents trained under CroMAC can obtain guaranteed lower bounds on state-action values to identify and choose the optimal action under a worst-case deviation when the received messages are perturbed. Concretely, we first model multi-agent communication as a multi-view problem, where every message stands for a view of the state. Then we extract a certificated joint message representation by a multi-view variational autoencoder (MVAE) that uses a product-of-experts inference network. For the optimization phase, we do perturbations in the latent space of the state for a certificate guarantee. Then the learned joint message representation is used to approximate the certificated state representation during training. Extensive experiments in several cooperative multi-agent benchmarks validate the effectiveness of the proposed CroMAC.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 65 of its 94 references that are in the map. First posted on arXiv 7 May 2023.
Builds on
65 of 94 references.
References · 65
- 2014Explaining and Harnessing Adversarial Examplescited by 7,364
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
- 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,057
- 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,558
- 2021The Surprising Effectiveness of PPO in Cooperative, Multi-Agent Gamescited by 942
- 2018QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learningcited by 708
- 2016Learning to Communicate with Deep Multi-Agent Reinforcement Learningcited by 575
- 2019Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithmscited by 567
- 2017Robust Adversarial Reinforcement Learningcited by 451
- 2016Learning Multiagent Communication with Backpropagationcited by 366
- 2018On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Modelscited by 328
- 2020QPLEX: Duplex Dueling Multi-Agent Q-Learningcited by 270
38 more
- 2021SMIL: Multimodal Learning with Severely Missing Modalitycited by 245
- 2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 238
- 2020Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Taskscited by 212
- 2018TarMAC: Targeted Multi-Agent Communicationcited by 208
- 2018Multimodal Generative Models for Scalable Weakly-Supervised Learningcited by 200
- 2022Multi-Agent Reinforcement Learning is a Sequence Modeling Problemcited by 155
- 2019A Review of Cooperative Multi-Agent Deep Reinforcement Learningcited by 147
- 2016A Survey of Multi-View Representation Learningcited by 107
- 2020Off-Policy Multi-Agent Decomposed Policy Gradientscited by 98
- 2019Dealing with Non-Stationarity in Multi-Agent Deep Reinforcement Learningcited by 93
- 2016Joint Multimodal Learning with Deep Generative Modelscited by 87
- 2021Scaling Multi-Agent Reinforcement Learning with Selective Parameter Sharingcited by 87
- 2019Learning Nearly Decomposable Value Functions Via Communication Minimizationcited by 86
- 2020Robust Deep Reinforcement Learning through Adversarial Losscited by 75
- 2021Learning Safe Multi-Agent Control with Decentralized Neural Barrier Certificatescited by 74
- 2014Generalized Product of Experts for Automatic and Principled Fusion of Gaussian Process Predictionscited by 67
- 2020Robust Reinforcement Learning using Adversarial Populationscited by 66
- 2021Learning by Aligning: Visible-Infrared Person Re-identification using Cross-Modal Correspondencescited by 66
- 2021ToM2C: Target-oriented Multi-agent Communication and Cooperation with Theory of Mindcited by 58
- 2019Risk Averse Robust Adversarial Reinforcement Learningcited by 57
- 2021Contrasting Centralized and Decentralized Critics in Multi-Agent Reinforcement Learningcited by 56
- 2020Learning Individually Inferred Communication for Multi-Agent Cooperationcited by 55
- 2021Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RLcited by 55
- 2021Multi-Agent Reinforcement Learning for Active Voltage Control on Power Distribution Networkscited by 54
- 2021Adversarial Attacks On Multi-Agent Communicationcited by 49
- 2022Towards a Standardised Performance Evaluation Protocol for Cooperative MARLcited by 49
- 2019Efficient Communication in Multi-Agent Reinforcement Learning via Variance Based Controlcited by 46
- 2022Efficient Adversarial Training without Attacking: Worst-Case-Aware Robust Reinforcement Learningcited by 43
- 2021CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothingcited by 42
- 2020On the Robustness of Cooperative Multi-Agent Reinforcement Learningcited by 39
- 2022Trustworthy Reinforcement Learning Against Intrinsic Vulnerabilities: Robustness, Safety, and Generalizabilitycited by 36
- 2022Multi-agent Dynamic Algorithm Configurationcited by 36
- 2019Learning Agent Communication under Limited Bandwidth by Message Pruningcited by 31
- 2021DRIBO: Robust Deep Reinforcement Learning via Multi-View Information Bottleneckcited by 31
- 2020Succinct and Robust Multi-Agent Communication With Temporal Message Controlcited by 29
- 2020The Emergence of Adversarial Communication in Multi-Agent Reinforcement Learningcited by 28
- 2020Towards Understanding Cooperative Multi-Agent Q-Learning with Value Factorizationcited by 26
- 2022Cooperative Multi-Agent Deep Reinforcement Learning for Reliable Surveillance via Autonomous Multi-UAV Controlcited by 26
50 most cited of 65 All 65 (JSON)
Built on by
Citing papers per year: 2025: 3.
3 papers.