Sequential Domain Adaptation by Synthesizing Distributionally Robust Experts
This work addresses domain adaptation for improving predictive accuracy in sequential target domains, but it appears incremental as it builds on existing robust optimization and aggregation methods.
The paper tackles the problem of poor prediction by least squares estimators in domain adaptation by synthesizing distributionally robust experts that are robust to moment conditions, and shows through numerical experiments on real data that these robust strategies can outperform non-robust interpolations.
Least squares estimators, when trained on a few target domain samples, may predict poorly. Supervised domain adaptation aims to improve the predictive accuracy by exploiting additional labeled training samples from a source distribution that is close to the target distribution. Given available data, we investigate novel strategies to synthesize a family of least squares estimator experts that are robust with regard to moment conditions. When these moment conditions are specified using Kullback-Leibler or Wasserstein-type divergences, we can find the robust estimators efficiently using convex optimization. We use the Bernstein online aggregation algorithm on the proposed family of robust experts to generate predictions for the sequential stream of target test samples. Numerical experiments on real data show that the robust strategies may outperform non-robust interpolations of the empirical least squares estimators.