2017年度第2回定例研究会「Artificial Intelligence Techniques Towards Adaptive Digital Games」(Marco Tamassia)

 下記の通り、客員研究員のTamassia氏による,今年度第2回定例研究会「Artificial Intelligence Techniques Towards Adaptive Digital Games」の開催いたします.みなさまふるってご参加ください。

日時
2017年7月12日(水)17:00~18:00
場所
立命館大学(びわこ・くさつキャンパス)クリエーションコア1階CRECORE-LA
アクセス
http://www.ritsumei.ac.jp/campusmap/bkc/
参加費
無料
事前申込み
不要
タイトル
Artificial Intelligence Techniques Towards Adaptive Digital Games
発表者
TAMASSIA Marco (Royal Melbourne Institute of Technology, Australia)
コーディネーター
THAWONMAS Ruck(立命館大学大学情報理工学部・立命館大学ゲーム研究センター)
Abstract:
Digital games rely on suspension of disbelief and challenge to immerse players in the game experience. Artificial Intelligence (AI) has a significant role in this task, both to provide adequate challenges to the player and to generate believable behaviors. An emerging area of application for digital games is augmented reality.
My work contributes to the field of AI in games, in particular by proposing techniques aimed at improving players experience. The technical contributions are in the field of learning from demonstration, abstraction in learning and dynamic difficulty adjustment. In particular, we propose a novel approach to learn options for the Options framework from demonstrations; a novel approach to handle progressively refined state abstractions for the Reinforcement Learning framework; a novel approach to dynamic difficulty adjustment based on state-action values, which in our experiments we compute via Monte Carlo Tree Search. The former two can help produce more powerful computer-controlled adversaries, while the latter can produce adversaries that play at the right level of difficulty. All these techniques have been tested in video games.
Among other things, these techniques could improve the quality of in-queue entertainment in theme parks. Village Roadshow, which sponsored part of my research and operates several theme parks in Australia, expressed interest in augmenting user experience via digital games. We developed an Augmented Reality game and measured perceived time in theme park customers that were queuing for an attraction, whether they were playing the game or not. Quantitative and qualitative analysis of the data suggests that the game helped reduce perceived time but could be improved. We speculate adding AI elements in the game can deliver some of the desired improvements.
共催
立命館大学情報理工学部知能エンターテインメント研究室 http://www.ice.ci.ritsumei.ac.jp/