报告人：Dr Simon Lucas 人工智能教授
In recent years there has been great progress in strong game-playing agents, with AI now superhuman on many challenging games such as Go, and becoming competitive on even harder games such as Starcraft. This has been achieved with a mix of Deep Reinforcement Learning and Statistical Forward Planning (SFP) algorithms. In this talk I’ll give a summary of some of the main achievements and current unsolved problems, as well as exploring opportunities for Game AI in other directions. This includes automatically tuning and playtesting games end evolving game content, and also how we can apply game AI to real-world decision making.