A transfer learning approach is presented to address the challenge of training video game agents with limited data. The approach decomposes games into objects, learns object models, and transfers models from known games to unfamiliar games to guide learning. Experiments show that the approach improves prediction accuracy over a comparable control, leading to more efficient exploration. Training of game agents is thus accelerated by transferring object models from previously learned games.

PDF

To Appear In Proceedings of the Twelfth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment 2016.

@inproceedings{braylan:aiide2016, title={Object-Model Transfer in the General Video Game Domain}, author={Alexander Braylan and Risto Miikkulainen}, booktitle={Proceedings of the Twelfth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment}, url="http://www.cs.utexas.edu/users/ai-lab?braylan:aiide16", year={2016} }