Practical Uncertainty for Decisions
Summary: Frameworks for incorporating model uncertainty into actionable decisions, with examples from public policy and infrastructure planning.
AI & Decision Science
Research and applied work on how models shape decisions under uncertainty — from probabilistic reasoning to algorithmic policy tools.
Approach
I focus on interpretable, robust methods: calibrating models to real-world signals, quantifying uncertainty, and designing decision rules that are transparent and testable.
Representative Work
- Uncertainty-aware ranking algorithms for resource allocation.
- Hybrid human-in-the-loop systems combining expert judgement with probabilistic models.
- Tooling for simulation-based policy evaluation under adversarial scenarios.