The room, named the Sports Book, is designed to replicate the scene and the vibe of a Las Vegas casino. The only major difference: Patrons could not place a bet on MLB, PGA or PBA competitions at the track. Only licensed venues in Nevada, Delaware, Oregon and Montana are permitted to accept wagers on those and other events drawing interest from the public.
Daily fantasy sports allow people to deposit money into accounts, create fantasy rosters of sports teams by selecting real players and compete against other contestants based on the statistical performances of those players to win money. Proponents say it's a game of skill, not chance, and shouldn't be regulated the way casinos are. But many states consider them to be forms of gambling.
Object-oriented representations in reinforcement learning have shown promise in transfer learning, with previous research introducing a propositional object-oriented framework that has provably efficient learning bounds with respect to sample complexity. However, this framework has limitations in terms of the classes of tasks it can efficiently learn. In this paper we introduce a novel deictic object-oriented framework that has provably efficient learning bounds and can solve a broader range of tasks. Additionally, we show that this framework is capable of zero-shot transfer of transition dynamics across tasks and demonstrate this empirically for the Taxi and Sokoban domains. Papers published at the Neural Information Processing Systems Conference.