|
Name of Speaker
|
Park Sinchaisri |
|
Schedule
|
10 Apr 2026, 10am – 11.30am (60 min talk + 30 min Q&A) |
|
Venue
|
BIZ1 0204
|
| Link to register
(via Zoom) |
|
|
Title
|
Algorithmic Advice, Human Compliance, and Learning
|
|
Abstract
|
Problem definition:Organizations increasingly deploy algorithmic tools to support complex operational decisions,raising a practical design question: how should these tools be built when designers care not only about immediate performance, butalso about preserving and building human skill that remains valuable when advice is unavailable, imperfect, or requires genuineoversight? We study how theprecisionof algorithmic advice shapes this trade-off.Methodology/results:We develop a stylized modelof advice-taking and learning. The model characterizes a reward-learning frontier: precise, action-level advice is easier to implementand improves payoffs while available through higher compliance, whereas broad, strategic advice requires interpretation, inducesgreater exploration, and generates knowledge that is portable, even when decision environments differ. We test the model’s predictionsin two online experiments in an electric-vehicle routing and charging task, representing typical characteristics of sequential decisiontasks. Consistent with the theory, precise numerical advice delivers the strongest gains during the advice phase, whereas broaderadvice can yield more robust performance after advice is removed, specifically if the new environment differs substantially, butnot completely. We use inverse reinforcement learning to recover interpretable latent objective components from action traces,distinguishing transient compliance from persistent internalization.Managerial implications:Our results provide design guidancefor advice systems that balance short-run operational efficiency with the development of long-run human capability. They also helpvalidate inverse reinforcement learning as an effective tool for estimating human behaviors in complex sequential tasks
|
|
About the Speaker
|
Park Sinchaisri is an Assistant Professor of Operations and IT Management at the Haas School of Business, University of California, Berkeley. His research draws on operations management, economics, machine learning, and behavioral science to study human decision-making in complex environments, design human-AI systems that improve decision-making, and develop strategies for managing the future of work. His work has been published in Management Science and Manufacturing & Service Operations Management, and has also appeared in leading human-computer interaction venues including CSCW. He received his PhD in Operations, Information and Decisions and an AM in Statistics from the Wharton School of the University of Pennsylvania, an SM in Computational Science and Engineering from MIT, and an ScB in Computer Engineering and Applied Mathematics-Economics from Brown University. Originally from Bangkok, Thailand, he hopes his research can help address urban challenges and improve outcomes for marginalized workers.
|
