Research map / Multi-agent systems and games / Cooperative multi-agent RL
MARS: A Meta-Adaptive Reinforcement Learning Framework for Risk-Aware Multi-Agent Portfolio Management
Jiayi Chen, Jing Li, Guiling Wang
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 the Meta-Adaptive Controller's dynamic weighting (vs fixed uniform weights) (ablation) | MARS · DJI 2024 | cumulative return (%): 17.10 -> 29.50 | Improved |
| add the Meta-Adaptive Controller's dynamic weighting (vs fixed uniform weights) (ablation) | MARS · DJI 2024 | Sharpe ratio: 1.71 -> 2.84 | Improved |
| add heterogeneous risk profiles to the ensemble (vs identical profiles) (ablation) | MARS · DJI 2024 | cumulative return (%): 22.21 -> 29.50 | Improved |
| add heterogeneous risk profiles to the ensemble (vs identical profiles) (ablation) | MARS · DJI 2024 | Sharpe ratio: 1.85 -> 2.84 | Improved |
| add heterogeneous risk profiles to the ensemble (vs identical profiles) (ablation) | MARS · DJI 2024 | maximum drawdown (%): -7.81 -> -5.39 | Improved |
| increase ensemble size from 5 to 10 agents (ablation) | MARS · DJI 2024 | cumulative return (%): 12.02 -> 29.50 | Improved |
| increase ensemble size from 15 to 10 agents (default N=10) (ablation) | MARS · DJI 2024 | cumulative return (%): 19.70 -> 29.50 | Hurt |
| MARS vs best baseline Sharpe ratio | MARS · DJI 2022 | Sharpe ratio: 70.6% relative improvement in Sharpe Ratio | Improved |
| MARS vs best baseline Sharpe ratio | MARS · DJI 2024 | Sharpe ratio: 101.4% relative improvement in Sharpe Ratio | Improved |
| MARS vs best baseline Sharpe ratio | MARS · HSI 2024 | Sharpe ratio: 35.5% relative improvement | Improved |
| MARS vs best baseline Sharpe ratio | MARS · HSI 2022 | Sharpe ratio: -0.66 vs -0.64 | No clear effect |
| MARS vs passive HSI index on return | MARS · HSI 2024 | cumulative return (%): 16.24 vs 24.46 | Hurt |
Closest papers by meaning
- Augmented Lagrangian Risk-constrained Reinforcement Learning for Portfolio Optimization (Student Abstract) (AAAI 2025 · similarity 0.87)
- Robust Reinforcement Learning in Finance: Modeling Market Impact with Elliptic Uncertainty Sets (NeurIPS 2025 · similarity 0.86)
- Risk-Conditioned Reinforcement Learning: A Generalized Approach for Adapting to Varying Risk Measures (AAAI 2024 · similarity 0.86)
- FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making (NeurIPS 2024 · similarity 0.85)
- Beyond CVaR: Leveraging Static Spectral Risk Measures for Enhanced Decision-Making in Distributional Reinforcement Learning (ICML 2025 · similarity 0.84)
- AlphaPareto: Formulaic Alpha Discovery with LLM-Guided Multi-Objective Reinforcement Learning (NeurIPS 2026 · similarity 0.84)
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.
