Research map / Multi-agent systems and games / Cooperative multi-agent RL
Secure Multi-agent Reinforcement Learning for Service Systems with Affinity and Byzantine Nodes: Stability Analysis and Protection Design
Yifan Jiang, Pan Jiasheng, Mengtian Li, Li Jin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add W-MSR aggregation and synchronized policy mixture under Byzantine attack | decentralized actor-critic MARL · LLM semantic routing | average service time per task: 4.46e+5 -> 6.88 | Improved |
| add W-MSR aggregation and synchronized policy mixture under Byzantine attack | decentralized actor-critic MARL · Multi-agent edge computing | average service time per task: 18.16 -> 4.13 | Improved |
| add W-MSR aggregation and synchronized policy mixture under Byzantine attack | decentralized actor-critic MARL · Smart manufacturing delivery | average service time per task: 36.28 -> 4.48 | Improved |
| add defense under Byzantine attack | decentralized actor-critic MARL · LLM semantic routing | learned affinity: 0.472 -> 0.664 | Improved |
| add defense under Byzantine attack | decentralized actor-critic MARL · Multi-agent edge computing | learned affinity: 0.351 -> 0.642 | Improved |
| add defense under Byzantine attack | decentralized actor-critic MARL · Smart manufacturing delivery | learned affinity: 0.409 -> 0.580 | Improved |
| add synchronized policy mixture on top of W-MSR (ablation) | resilient MARL · LLM, MEC, delivery | queue stability | Improved |
| add W-MSR on top of policy mixture (ablation) | resilient MARL · LLM, MEC, delivery | learning quality | Improved |
| use proposed method versus mean, median, projection aggregation | resilient MARL · MEC | performance under Byzantine attack | Improved |
| use proposed method versus robust aggregation baselines | resilient MARL · LLM routing | performance under Byzantine attack | No clear effect |
| use proposed method versus robust aggregation baselines | resilient MARL · Smart manufacturing delivery | performance under Byzantine attack | No clear effect |
Closest papers by meaning
- Byzantine Robust Cooperative Multi-Agent Reinforcement Learning as a Bayesian Game (ICLR 2024 · similarity 0.88)
- Learning Distributed and Fair Policies for Network Load Balancing as Markov Potential Game (NeurIPS 2022 · similarity 0.87)
- Bayesian Ego-graph inference for Networked Multi-Agent Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- Scalable Constrained Policy Optimization for Safe Multi-agent Reinforcement Learning (NeurIPS 2024 · similarity 0.86)
- Exploring the Efficacy of Multi-Agent Reinforcement Learning for Autonomous Cyber Defence: A CAGE Challenge 4 Perspective (AAAI 2025 · similarity 0.86)
- Beyond Training Time, Test-Time Coordination is Essential for Cooperative MARL (NeurIPS 2026 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
