1 October 2018: Kobi Gal Abstract Advances in technology and interface design are enabling heterogenous group activities that include both human and computer participants. The need to support these highly diverse interactions brings new and significant challenges to AI; how to design efficient representations for describing online group interactions; how to design incentives that keep participants motivated and productive; and how to provide useful, non-intrusive information to the system designers and to users to explain group behavior and to help them decide whether and how to intervene. I will describe ongoing work in my lab that addresses these challenges using novel applications of reinforcement learning and probabilistic modeling. I will demonstrate t these approaches in two projects: enhancing volunteer engagement in citizen science, and modeling students' activities in mixed reality simulations. Relevant papers Avi Segal, Kobi Gal, Ece Kamar, Eric Horvitz, Grant Miller. Optimizing Interventions via Offline Policy Evaluation: Studies in Citizen Science. National Conference on Artificial Intelligence (AAAI), New Orleans, LA, February, 2018. Nicholas Hoernle, Kobi Gal, Barbara Grosz, Pavlos Protopapas, and Andee Rubin. Modeling the Effects of Students’ Interactions with Immersive Simulations using Markov Switching Systems. Educational Data Mining (EDM), Buffalo, NY, 2018 Oct 01 2018 14.00 - 15.00 1 October 2018: Kobi Gal Advances in Human-Informed Decision-Making: Representations, Inference, and experiments IF 4.31/4.33
1 October 2018: Kobi Gal Abstract Advances in technology and interface design are enabling heterogenous group activities that include both human and computer participants. The need to support these highly diverse interactions brings new and significant challenges to AI; how to design efficient representations for describing online group interactions; how to design incentives that keep participants motivated and productive; and how to provide useful, non-intrusive information to the system designers and to users to explain group behavior and to help them decide whether and how to intervene. I will describe ongoing work in my lab that addresses these challenges using novel applications of reinforcement learning and probabilistic modeling. I will demonstrate t these approaches in two projects: enhancing volunteer engagement in citizen science, and modeling students' activities in mixed reality simulations. Relevant papers Avi Segal, Kobi Gal, Ece Kamar, Eric Horvitz, Grant Miller. Optimizing Interventions via Offline Policy Evaluation: Studies in Citizen Science. National Conference on Artificial Intelligence (AAAI), New Orleans, LA, February, 2018. Nicholas Hoernle, Kobi Gal, Barbara Grosz, Pavlos Protopapas, and Andee Rubin. Modeling the Effects of Students’ Interactions with Immersive Simulations using Markov Switching Systems. Educational Data Mining (EDM), Buffalo, NY, 2018 Oct 01 2018 14.00 - 15.00 1 October 2018: Kobi Gal Advances in Human-Informed Decision-Making: Representations, Inference, and experiments IF 4.31/4.33
Oct 01 2018 14.00 - 15.00 1 October 2018: Kobi Gal Advances in Human-Informed Decision-Making: Representations, Inference, and experiments
