Research map / Multi-agent systems and games / Cooperative multi-agent RL
Scalable Multi-Agent Reinforcement Learning for Networked Systems with Average Reward
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Bound each agent's total interaction strength below 1 | Networked MARL with average reward · Average-reward networked MARL | exponential decay bound: the $(\frac{\bar{r}}{1-\rho},\rho)$ exponential decay property holds | Helped |
| Truncate each agent's Q-function to its k-hop neighborhood | Scalable Actor-Critic (SAC) · Average-reward networked MARL | policy-gradient approximation error: $\Vert \hat{h}_i(\theta) - \nabla_{\theta_i} J(\theta)\Vert \leq cL_i \rhok $ | Helped |
| Use the two-time-scale actor-critic with truncated Q-functions | Scalable Actor-Critic (SAC) · Average-reward networked MARL | stationarity gradient norm: \liminf_{T\rightarrow\infty} \Vert \nabla J(\theta(T))\Vert \leq L \frac{ c \rhok}{1-\mu_{D}} | Helped |
| Apply the bounded-interaction condition to the discounted-reward setting | Networked MARL · Discounted-reward networked MARL | exponential decay rate: the $(\frac{\bar{r}}{1 - \gamma\rho},\rho\gamma) $ exponential decay property holds | Helped |
Closest papers by meaning
- Multi-Agent Reinforcement Learning in Stochastic Networked Systems (NeurIPS 2021 · similarity 0.95)
- Bayesian Ego-graph inference for Networked Multi-Agent Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
- Finite-Time Global Optimality Convergence in Deep Neural Actor-Critic Methods for Decentralized Multi-Agent Reinforcement Learning (ICML 2025 · similarity 0.88)
- A Sharper Global Convergence Analysis for Average Reward Reinforcement Learning via an Actor-Critic Approach (ICML 2025 · similarity 0.87)
- Bridging Training and Execution via Dynamic Directed Graph-Based Communication in Cooperative Multi-Agent Systems (AAAI 2025 · similarity 0.87)
- Scalable Constrained Policy Optimization for Safe Multi-agent Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
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.