Extends multi-headed neural networks for mixed-strategy deviation payoffs by adding environment parameters as input dimensions, enabling a single model to generalize across families of games with better accuracy and less data.
ACM Transactions on Internet Technology 15, 1 (2015), 3:1–3:41
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.GT 1years
2023 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning Parameterized Families of Games
Extends multi-headed neural networks for mixed-strategy deviation payoffs by adding environment parameters as input dimensions, enabling a single model to generalize across families of games with better accuracy and less data.