Market Regime Council for Dynamic Credit Assignment in Multi-Agent LLM Decision Systems
Researchers have developed a new multi-agent LLM system called Market Regime Council (MRC) designed for dynamic credit assignment in portfolio management. MRC computes exact Shapley credits across agent coalitions to determine agent weighting and improve transparency. In simulations over 1,037 trading days across 13 crypto assets, MRC achieved a Sharpe ratio of 1.51 and a cumulative return of 440.1%, outperforming baseline methods. AI
IMPACT Introduces a novel method for credit assignment in multi-agent LLM systems, potentially improving performance and transparency in financial applications.