Timezone: »

Evaluating Robustness to Dataset Shift via Parametric Robustness Sets
Nikolaj Thams · Michael Oberst · David Sontag

We give a method for proactively identifying small, plausible shifts in distribution which lead to large differences in model performance. To ensure that these shifts are plausible, we parameterize them in terms of interpretable changes in causal mechanisms of observed variables. This defines a parametric robustness set of plausible distributions and a corresponding worst-case loss. We construct a local approximation to the loss under shift, and show that problem of finding worst-case shifts can be efficiently solved.

Author Information

Nikolaj Thams (University of Copenhagen)
Michael Oberst (MIT)
David Sontag (Massachusetts Institute of Technology)

More from the Same Authors