Research

Methods and evidence for deciding when machine-learning systems can be trusted under changing operating conditions.

My research addresses a basic deployment question: when should a machine-learning system no longer be trusted? I approach this question through three connected lines of work.

Distribution shift and monitoring §

Machine-learning models can fail when their operating conditions differ from their training data, often while remaining highly confident. I develop out-of-distribution (OOD) detection methods that help recognize inputs and situations for which a model’s predictions may be unreliable.

Representative work:

Reasoning with structured knowledge §

Unusual situations are not always visually or statistically unusual. My neuro-symbolic work combines learned representations with logical, probabilistic, or natural-language knowledge about a domain. This makes it possible to detect semantic inconsistencies and gives people a way to inspect or revise some of the assumptions used by a detector.

Representative work:

Evaluation and evidence §

A detector is only useful if its evaluation supports clear, reproducible claims. I work on software and evaluation protocols that make OOD experiments easier to compare, reproduce, and interpret.

Representative work:

Application context §

In a safety-critical perception system, a change in operating conditions can make a learned component unreliable. Monitoring provides evidence that can trigger an operational response such as abstention, human review, or a fallback mode:

Operating-condition change Unreliable perception Monitoring evidence Abstention or fallback

My longer-term objective is to understand how such model-level evidence can support defensible decisions about deploying machine learning in systems such as rail transport and autonomous vehicles.