I'm working on artificial intelligence for making games more fun to play, and on human-computer interaction techniques to make academic game AI more accessible for non-experts.
My current research focuses on creating aligned agent behaviors. One open challenge in this field is creating effective communication channels between the agent and the user. From the user agent, this channel has to enable it to explain its decision procedure to the user. From the user standpoint, communication has to provide her with the tools to give precise feedback to the agent. My collaborators at the Igarashi Lab, and I tackle this challenge by designing human-computer interaction methods that minimize the needed human feedback, and reinforcement learning based agents that explain their thinking process.
Christian Arzate Cruz and Takeo Igarashi."Interactive Explanations: Diagnosis and Repair of Reinforcement Learning Based Agent Behaviors" In IEEE Conference on Games (CoG), 2021. Project Page
- Christian Arzate Cruz and Takeo Igarashi."MarioMix: Creating Aligned Playstyles for Bots with Interactive Reinforcement Learning" In Extended Abstracts of the Annual Symposium on Computer-Human Interaction in Play (CHI Play), 2020 https://doi.org/10.1145/3383668.3419938. Project Page
- Christian Arzate Cruz and Takeo Igarashi."A Survey on Interactive Reinforcement Learning: Design Principles and Open Challenges" In Proceedings of the 2020 on Designing Interactive Systems Conference (DIS), 2020 http://doi.org/10.1145/3357236.3395525.
- Hirotaka Suetake, Tsukasa Fukusato, Christian Arzate Cruz, Andy Nealen, and Takeo Igarashi. "Interactive Design Exploration of Game Stages Using Adjustable Synthetic Testers" In Proceedings of the 15th International Conference on the Foundations of Digital Games (FDG), 2020 https://doi.org/10.1145/3402942.3402982.
- Hirotaka Suetake，Tsukasa Fukusato，Christian Arzate Cruz，Andy Nealen，and Takeo Igarashi. "ゲームAIによるパラメータ空間解析を活用した難易度調整用インターフェース" The 27th Workshop on Interactive Systems and Software (WISS), 2019.
- Christian Arzate Cruz and Jorge Adolfo Ramirez Uresti. "HRLB⌃2: A Game AI Architecture for Believable Bots That Unifies the Elements of Flow and Reinforcement Learning" PhD Thesis, 2018.
- Christian Arzate Cruz and Jorge Adolfo Ramirez Uresti. "HRLB⌃2: A Reinforcement Learning Based Framework for Believable Bots" Applied Sciences, 2018 https://doi.org/10.3390/app8122453.
- Christian Arzate Cruz and Jorge Adolfo Ramirez Uresti. "Player-centered game AI from a flow perspective: Towards a better understanding of past trends and future directions." Entertainment Computing, 2017 https://doi.org/10.1016/j.entcom.2017.02.003.
- Christian Arzate Cruz and Jorge Adolfo Ramirez Uresti. "Personalized AI for Groups of Non-PlayerCharacters." In Doctoral Consortium of the Mexican International Conference on Artificial Intelligence (MICAI), 2013. PDF
- Christian Arzate Cruz, Isaac Rudomin Goldberg, and Jorge Adolfo Ramirez Uresti. "Método Robusto para Detectar Dedos Usando Profundidad." In Congreso Mexicano de Inteligencia Artificial (COMIA), 2013. PDF