Domain Adaptation with Asymmetrically-Relaxed Distribution Alignment
Domain adaptation addresses the common problem when the target distribution generating our test data drifts from the source (training) distribution. While absent assumptions, domain adaptation is impossible, strict conditions, e.g. covariate or label shift, enable principled algorithms. Recently-proposed domain-adversarial approaches consist of aligning source and target encodings, often motivating this approach as minimizing two (of three) terms in a theoretical bound on target error. Unfortunately, this minimization can cause arbitrary increases in the third term, e.g. they can break down under shifting label distributions. We propose asymmetrically-relaxed distribution alignment, a new approach that overcomes some limitations of standard domain-adversarial algorithms. Moreover, we characterize precise assumptions under which our algorithm is theoretically principled and demonstrate empirical benefits on both synthetic and real datasets.
Code (1)
Tasks
Domain AdaptationSimilar Papers 제목 키워드 기반
Connecting sufficient conditions for domain adaptation: source-guided uncertainty, relaxed divergences and discrepancy localization
Recent advances in domain adaptation establish that requiring a low risk on the source domain and equal feature marginals degrade the adaptation's performance. At the same time, empirical evidence shows that incorporatin…
Domain AdaptationAdversarial Support Alignment
We study the problem of aligning the supports of distributions. Compared to the existing work on distribution alignment, support alignment does not require the densities to be matched. We propose symmetric support differ…
Domain AdaptationRLSbench: Domain Adaptation Under Relaxed Label Shift
Despite the emergence of principled methods for domain adaptation under label shift, their sensitivity to shifts in class conditional distributions is precariously under explored. Meanwhile, popular deep domain adaptatio…
Domain AdaptationA Turn Toward Better Alignment: Few-Shot Generative Adaptation with Equivariant Feature Rotation
Few-shot image generation aims to effectively adapt a source generative model to a target domain using very few training images. Most existing approaches introduce consistency constraints-typically through instance-level…
Image GenerationMOT: Masked Optimal Transport for Partial Domain Adaptation
As an important methodology to measure distribution discrepancy, optimal transport (OT) has been successfully applied to learn generalizable visual models under changing environments. However, there are still limitat…
Domain AdaptationPartial Domain Adaptation