Dynamic Resource Allocation for Ensemble Determinization MCTS

This paper proposes enhancements for Ensemble Determinization Monte Carlo Tree Search (MCTS) by introducing two axes for dynamic resource allocation: Dynamic Number of Determinizations and Dynamic Simulation Allocation.

• Dynamic Number of Determinizations adjusts the count of determinization trees based on search behavior.
• Dynamic Simulation Allocation splits the simulation budget nonuniformly across trees to maximize knowledge gain.
• The method was tested in Jaipur, Lost Cities, and Splendor using iteration- and time-based settings.

These configurations yield a statistically significant increase in the algorithm's strength.