Print Email Facebook Twitter Inductive Program Synthesis through using Monte Carlo Tree Search guided by a heuristic-based loss function Title Inductive Program Synthesis through using Monte Carlo Tree Search guided by a heuristic-based loss function Author Matulewicz, Nadia (TU Delft Electrical Engineering, Mathematics and Computer Science; TU Delft Algorithmics) Contributor Dumančić, S. (mentor) Degree granting institution Delft University of Technology Programme Computer Science and Engineering Project CSE3000 Research Project Date 2022-01-28 Abstract Recently, a new and promising Inductive Program Synthesis (IPS) system, Brute, showed the potential of using a heuristic-based loss function. However, Brute also has its limitations and struggles with escaping local optima. The Monte Carlo Tree Search might offer a solution to this problem since it balances between exploitation and exploration. I design MUTE, a new IPS system which uses MCTS guided by a heuristic-based loss function. MUTE's performance is tested and compared to other IPS systems in three diverse domains, namely robot planning, ASCII art and string transformations. MUTE's performance for string transformations is a first indication that MUTE can outperform Brute and other IPS systems. Manual analysis of the results shows that MUTE can indeed escape local optima. Two branch reducing enhancements, namely the removal of similar programs and the removal of tokens that show no potential, are essential for the success of MUTE. Subject Inductive Program SynthesisMUTEBruteMonte Carlo Tree SearchUCTProgram Synthesislocal optima To reference this document use: http://resolver.tudelft.nl/uuid:4ad6c3bf-6a66-4666-8700-aecd06b4407d Coordinates 52.0022, 4.3736 Bibliographical note https://github.com/victorvwier/BEP_project_synthesis Implementation code of MUTE and IPS systems that were used for comparison. Also contains uses test and training data. Part of collection Student theses Document type bachelor thesis Rights © 2022 Nadia Matulewicz Files PDF combinepdf_6_.pdf 564.11 KB Close viewer /islandora/object/uuid:4ad6c3bf-6a66-4666-8700-aecd06b4407d/datastream/OBJ/view