Research Program 5 (R5)

Interactive AI

The goal of FCAI’s research program Interactive AI is to enable AI that people can naturally work and solve problems with, and which demonstrates the ability to better understand our goals and abilities, takes initiative more sensitively, aligns its objectives with us, and supports us. This research program contributes to FCAI research objective Understandability (objective 3) by developing methods for collaborative forms of AI: the ability to infer human beliefs and abilities from observations and predicting the consequences of its actions on humans.

Our main expertise is in Bayesian and reinforcement learning (RL) based models of foundational capabilities in AI-human interaction, such as, among others, theory of mind (AI infers beliefs and mental causes of human actions), causal prediction (AI predicts the consequences of its actions on humans), and value-alignment (AI adapts to inferred motivations and needs of humans). The outcomes of Interactive AI research program include generative models of human interactive behavior based on the Bayesian brain hypothesis, inference of cognitive models from data using approximate Bayesian computation, interactive modeling of user intent through RL, and RL methods for sample-efficient policy search in robotics and computational agents.

Examples of publications:

• Hee-Seung Moon, Antti Oulasvirta, Byungjoo Lee. 2023. Amortized Inference with User Simulations. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (CHI 2023), to appear.

• John H. Williamson, Antti Oulasvirta, Per Ola Kristensson, Nikola Banovic (Eds.). 2022. Bayesian Methods for Interaction and Design. Cambridge University Press.

• Noshaba Cheema, Laura A Frey-Law, Kourosh Naderi, Jaakko Lehtinen, Philipp Slusallek, Perttu Hämäläinen. 2020. Predicting Mid-Air Interaction Movements and Fatigue Using Deep Reinforcement Learning. Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems

• Anna Maria Feit, Mathieu Nancel, Maximilian John, Andreas Karrenbauer, Daryl Weir, Antti Oulasvirta. 2021. AZERTY améliore: Computational design on a national scale. Communications of the ACM.

Program poster (pdf)

Coordinating professor: Antti Oulasvirta – antti.oulasvirta at aalto.fi

People

The groups of following PIs take part in the Research Program Interactive AI. If you would like to join this program, please contact the coordinating professor.

• Miguel Bordallo, VTT

• Dorota Glowacka, University of Helsinki

• Christian Guckelsberger, Aalto University

• Perttu Hämäläinen, Aalto University

• Giulio Jacucci, University of Helsinki

• Samuel Kaski, Aalto University

• Arto Klami, University of Helsinki

• Mikko Kurimo, Aalto University

• Ville Kyrki, Aalto University

• Antti Oulasvirta, Aalto University – coordinating professor

• Kai Puolamäki, University of Helsinki

• Tuukka Ruotsalo, University of Helsinki

• Fabrice Saffre, VTT

• Caj Södergård, VTT

• Aki Vehtari, Aalto University

• Robin Welsch, Aalto University

• Yu Xiao, Aalto University