[2305.09840] Scale-Adaptive Balancing of Exploration and Exploitation in Classical Planning

Computer Science > Artificial Intelligence arXiv:2305.09840 (cs) [Submitted on 16 May 2023 (v1), last revised 26 Mar 2026 (this version, v4)] Title:Scale-Adaptive Balancing of Exploration and Exploitation in Classical Planning Authors:Stephen Wissow, Masataro Asai View a PDF of the paper titled Scale-Adaptive Balancing of Exploration and Exploitation in Classical Planning, by Stephen Wissow and 1 other authors View PDF Abstract:Balancing exploration and exploitation has been an important problem in both game tree search and automated planning. However, while the problem has been extensively analyzed within the Multi-Armed Bandit (MAB) literature, the planning community has had limited success when attempting to apply those results. We show that a more detailed theoretical understanding of MAB literature helps improve existing planning algorithms that are based on Monte Carlo Tree Search (MCTS) / Trial Based Heuristic Tree Search (THTS). In particular, THTS uses UCB1 MAB algorithms in an ad hoc manner, as UCB1's theoretical requirement of fixed bounded support reward distributions is not satisfied within heuristic search for classical planning. The core issue lies in UCB1's lack of adaptations to the different scales of the rewards. We propose GreedyUCT-Normal, a MCTS/THTS algorithm with UCB1-Normal bandit for agile classical planning, which handles distributions with different scales by taking the reward variance into consideration, and resulted in an improved algorithmic ...