Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
Improving Human-algorithm collaboration: Causes and Mitigation of Over- and Under-Adherence
Maya Balakrishnan1, Kris Ferreira1, Jordan Tong2
1Harvard Business School, United States of America; 2Wisconsin School of Business, United States of America
Even if algorithms make better predictions than humans on average, humans may sometimes have private information which an algorithm does not have access to, which can improve performance. When deciding how to use and adjust an algorithm’s recommendations we hypothesize people are biased towards a predictable heuristic leading to over-adhering to the algorithm’s predictions when their private information is valuable and under-adhering when it is not. We test these results in two lab experiments.
Human-centric AI for sequential decision-making: a case study on electric vehicle charging
Philippe Blaettchen1, Park Sinchaisri2
1City, University of London; 2University of California, Berkeley
We develop a sequential decision-making task in the form of a virtual electric vehicle driving game in which the participant needs to make sequential charging decisions when facing uncertain traffic and receiving machine-generated recommendations on their strategy. Our experimental results offer key insights into how humans make decisions and respond to those recommendations, allowing us to design a better human-centric recommendation system.
Multi-treatment forest approach for analyzing the heterogeneous effects of team familiarity
Minmin Zhang1, Guihua Wang1, Wally Hopp2, Michael Mathis2
1University of Texas Dallas, USA; 2University of Michigan Ann Arbor, USA
We examine the effect of team familiarity on surgery duration. We develop a new approach, which we call the “MT forest” approach, to estimate heterogeneous effects of multiple treatments. We find (1) an increase in team familiarity score significantly reduces surgery duration, and (2) the effect of team familiarity is heterogeneous across patients with different features. Finally, we develop an optimization model to better match surgical teams with patients.
