Name
Abstract | ||
|---|---|---|
The game of Hex can be played on multiple boardsizes. Transferring neural net knowledge learned on one boardsize to other boardsizes is of interest, since deep neural nets usually require large size of high quality data to train, whereas expert games can be unavailable or difficult to generate. In this paper we investigate neural transfer learning in Hex. We show that when only boardsize independent neurons are used, the resulting neural net obtained from training on one base boardsize can effectively generalize - without fine-tuning - to multiple target boardsizes, larger or smaller. When transferring to larger boardsizes, fine-tuning provides faster learning and better performance. The strength of the transferable network can be amplified with search: with a single neural net model trained on games from a base boardsize, we obtain players stronger than MoHex 2.0 on multiple target boardsizes. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.3233/ICG-180055 | ICGA JOURNAL |
Keywords | Field | DocType |
Hex game,deep learning,transfer learning,neural network | Computer science,Artificial intelligence,Artificial neural network | Journal |
Volume | Issue | ISSN |
40 | 3 | 1389-6911 |
Citations | PageRank | References |
0 | 0.34 | 16 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Chao Gao | 1 | 42 | 5.78 |
| Siqi Yan | 2 | 0 | 0.34 |
| Ryan B. Hayward | 3 | 312 | 44.97 |
| Martin Müller | 4 | 549 | 68.48 |