Re-determinizing MCTS in Hanabi
James Goodman · 2019 IEEE Conference on Games (CoG) · 2019
We introduce Re-determinizing IS-MCTS, a novel extension of Information Set Monte Carlo Tree Search (IS-MCTS) that prevents a leakage of hidden information into opponent models that can occur in MCTS, and is particularly severe in Hanabi. When combined with a learned evaluation function to estimate leaf node values and avoid full simulations during MCTS this won the Hanabi competition at the Computation Intelligence in Games (CIG) 2018.