Summary of Scale-adaptive Balancing Of Exploration and Exploitation in Classical Planning, by Stephen Wissow et al.
Scale-Adaptive Balancing of Exploration and Exploitation in Classical Planning
by Stephen Wissow, Masataro Asai
First submitted to arxiv on: 16 May 2023
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a new algorithm for balancing exploration and exploitation in automated planning, which is an important problem in game tree search. The authors draw inspiration from the Multi-Armed Bandit (MAB) literature, but adapt it to suit the needs of classical planning. Specifically, they develop GreedyUCT-Normal, a Monte Carlo Tree Search (MCTS)/Trial Based Heuristic Tree Search (THTS) algorithm that incorporates UCB1-Normal bandit. This approach handles distributions with different scales by considering reward variance, resulting in improved performance compared to existing algorithms. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The new algorithm helps find more plans with fewer node expansions, making it a useful tool for agile classical planning. The authors show that their approach outperforms Greedy Best First Search and existing MCTS/THTS-based algorithms. |
