Thursday, August 3, 2023 - 10:00am
Event Calendar Category
Other LIDS Events
Speaker Name
Serena Wang
Affiliation
University of California, Berkeley
Building and Room Number
LIDS Lounge
From the social sciences to machine learning, it has been well documented that metrics to be optimized are not always aligned with social welfare. In healthcare, Dranove et al. (2003) showed that publishing surgery mortality metrics actually harmed the welfare of sicker patients by increasing provider selection behavior. Using a principal-agent model, we directly study the incentive misalignments that arise from such average treated outcome metrics, and show that the incentives driving treatment decisions would align with maximizing total patient welfare if the metrics (i) accounted for counterfactual untreated outcomes and (ii) considered total welfare instead of average welfare among treated patients. Operationalizing this, we show how counterfactual metrics can be modified to satisfy desirable properties when used for ranking. Extending to realistic settings when the providers observe more about patients than the regulatory agencies do, we bound the decay in performance by the degree of information asymmetry between the principal and the agent. In doing so, our model connects principal-agent information asymmetry with unobserved heterogeneity in causal inference.
Serena is a final-year PhD student in Computer Science at University of California, Berkeley. supported by the NSF Graduate Research Fellowship and the Apple Scholars in AI/ML PhD fellowship. She has also concurrently worked at Google Research at 20% time for the last six years, where she is part of the Discrete Algorithms Group with Ravi Kumar and have also worked with Maya Gupta on the Glassbox Machine Learning Research team. Serena’s research focuses on understanding and improving the long-term societal impacts of machine learning by rethinking ML algorithms and their surrounding incentives and practices. Serena is particularly interested in gaps between metrics and goals, and how those gaps may be bridged through algorithmic improvements, analysis of multi-agent incentives, and better qualitative understanding of problem formulation. Serena employs tools from robust optimization, statistics, learning theory, and most recently, microeconomics.