Choosing Well Your Opponents: How to Guide the Synthesis of Programmatic Strategies
Authors: Rubens O. Moraes, David S. Aleixo, Lucas N. Ferreira, Levi H. S. Lelis
IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We empirically demonstrate the advantages of our approach while guiding a local search algorithm for synthesizing strategies in three games, including Micro RTS, a challenging real-time strategy game. |
| Researcher Affiliation | Academia | Rubens O. Moraes1,2 , David S. Aleixo1 , Lucas N. Ferreira2 and Levi H. S. Lelis2 1 Departamento de Inform atica, Universidade Federal de Vic osa, Brazil 2Department of Computing Science, University of Alberta, Canada Alberta Machine Intelligence Institute (Amii) |
| Pseudocode | Yes | Algorithm 1 Programmatic PSRO |
| Open Source Code | Yes | Our code is at https://github.com/rubensolv/Local Learner IJCAI |
| Open Datasets | Yes | We evaluate 2L on three two-player zero-sum games: Micro RTS [Onta n on et al., 2018], Poachers & Rangers, and Climbing Monkeys.1https://github.com/Farama-Foundation/Micro RTS/wiki |
| Dataset Splits | No | The paper describes how games are played and evaluated (e.g., number of games, winning rates, tournament structure) but does not provide specific train/validation/test dataset splits like percentages or sample counts. |
| Hardware Specification | No | The research was carried out using computational resources from Compute Canada. |
| Software Dependencies | No | The paper mentions using 'Medeiros et al. [2022] s DSL for Micro RTS' but does not provide specific software dependencies or version numbers for any libraries, frameworks, or tools used in the experiments. |
| Experiment Setup | No | The paper describes the general algorithm (Local Learner) and its interaction with the Hill Climbing search, including mutation operations and initialization. However, it does not provide specific numerical hyperparameters (e.g., learning rates, batch sizes, number of epochs) or other concrete system-level training settings typically found in an experimental setup. |