Research map / Multi-agent systems and games / Cooperative multi-agent RL
Adaptive Regulation via Dual-Layer Evolution (ARDE): A Multi-Agent Approach to Balancing Efficiency, Fairness, and Diversity in Crowdsourced Platforms
XuWen Zhang, Xiao Xue, Xia Xie, Qun Ma
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 ARDE dual-layer governance to individual Q-learning | courier Q-learning agents · LaDe-D delivery simulation | Gini change (%): +27.9 -> -1.3 | Improved |
| add ARDE governance to individual Q-learning | courier Q-learning agents · LaDe-D delivery simulation | order completion rate: 0.189 -> 0.411 | Improved |
| add ARDE governance to individual Q-learning | courier Q-learning agents · LaDe-D delivery simulation | strategy entropy: 0.365 -> 0.997 | Improved |
| add LLM-guided adaptive outer agent on top of reward shaping (ablation) | reward shaping · LaDe-D delivery simulation | strategy entropy: 1.029 -> 0.997 | Hurt |
| add LLM-guided adaptive outer agent on top of reward shaping (ablation) | reward shaping · LaDe-D delivery simulation | Gini coefficient: 0.329 -> 0.339 | Hurt |
| add LLM-guided adaptive outer agent on top of reward shaping (ablation) | reward shaping · LaDe-D delivery simulation | Gini change (%): +1.2 -> -1.3 | Improved |
| add ARDE adaptive governance over a static rule (ablation) | static rule · LaDe-D delivery simulation | Gini change (%): +20.9 -> -1.3 | Improved |
| use ARDE governance over rule-based control | rule-based control · LaDe-D delivery simulation | order completion rate: 0.418 -> 0.411 | No clear effect |
| use ARDE-PPO instead of MAPPO under identical inner learner (ablation) | PPO inner learner · LaDe-D delivery simulation | Gini change (%): +2.1 -> -1.1 | Improved |
| use ARDE-PPO instead of MAPPO under identical inner learner (ablation) | PPO inner learner · LaDe-D delivery simulation | strategy entropy: 1.012 -> 1.005 | Hurt |
Closest papers by meaning
- Economy of Minds: Emerging Multi-Agent Intelligence with Economic Interactions (NeurIPS 2026 · similarity 0.85)
- Steering the Herd: A Framework for LLM-based Control of Social Learning (ICLR 2026 · similarity 0.84)
- Fair Cooperation in Mixed-Motive Games via Conflict-Aware Gradient Adjustment (NeurIPS 2025 · similarity 0.84)
- Network Intervention by Polling Strategic Agents (NeurIPS 2026 · similarity 0.84)
- One Step is Enough: Multi-Agent Reinforcement Learning Based on One-Step Policy Optimization for Order Dispatch on Ride-Sharing Platforms (NeurIPS 2026 · similarity 0.83)
- A Statistical Framework for Algorithmic Collective Action with Multiple Collectives (NeurIPS 2026 · similarity 0.83)
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.
