Szczegóły publikacji
Opis bibliograficzny
The application of Co-evolutionary genetic programming and TD(1) reinforcement learning in large-scale strategy game VCMI / Łukasz Wilisowski, Rafał DREŻEWSKI // W: Agent and multi-agent systems: technologies and applications : 9th KES international conference, KES-AMSTA 2015 : Sorrento, Italy, June 17–19, 2015 : proceedings / eds Gordan Jezic, Robert J. Howlett, Lakhmi C. Jain. — Cham [etc.] : Springer International Publishing, 2015. — (Smart Innovation, Systems and Technologies ; ISSN 2190-3018 ; vol. 38). — ISBN: 978-3-319-19727-2; e-ISBN: 978-3-319-19728-9. — S. 81–93. — Bibliogr.
Autorzy (2)
- Wilisowski Łukasz
- AGHDreżewski Rafał
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 90285 |
|---|---|
| Data dodania do BaDAP | 2015-07-15 |
| DOI | 10.1007/978-3-319-19728-9_7 |
| Rok publikacji | 2015 |
| Typ publikacji | materiały konferencyjne (aut.) |
| Otwarty dostęp |  |
| Konferencja | 9th KES International Conference on Agent and Multi-Agent Systems -Technologies and Applications |
| Czasopismo/seria | Smart Innovation, Systems and Technologies |
Abstract
VCMI is a new, open-source project that could become one of the biggest testing platform for modern AI algorithms in the future. Its complex environment and turn-based gameplay make it a perfect system for any AI driven solution. It also has a large community of active players which improves the testability of target algorithms. This paper explores VCMI's environment and tries to assess its complexity by providing a base solution for battle handling problem using two global optimization algorithms: Co-Evolution of Genetic Programming Trees and TD(1) algorithm with Back Propagation neural network. Both algorithms have been used in VCMI to evolve battle strategies through a fully autonomous learning process. Finally, the obtained strategies have been tested against existing solutions and compared with players' best tactics.
