LGAIOct 26, 2021

Mapping conditional distributions for domain adaptation under generalized target shift

arXiv:2110.15057v226 citations
Originality Highly original
AI Analysis

This addresses a challenging domain adaptation problem for machine learning applications where conditional and label shifts occur, offering a flexible and scalable solution.

The paper tackles unsupervised domain adaptation under generalized target shift by learning an optimal transport map to align pretrained representations, achieving strong theoretical guarantees and outperforming state-of-the-art methods on several datasets.

We consider the problem of unsupervised domain adaptation (UDA) between a source and a target domain under conditional and label shift a.k.a Generalized Target Shift (GeTarS). Unlike simpler UDA settings, few works have addressed this challenging problem. Recent approaches learn domain-invariant representations, yet they have practical limitations and rely on strong assumptions that may not hold in practice. In this paper, we explore a novel and general approach to align pretrained representations, which circumvents existing drawbacks. Instead of constraining representation invariance, it learns an optimal transport map, implemented as a NN, which maps source representations onto target ones. Our approach is flexible and scalable, it preserves the problem's structure and it has strong theoretical guarantees under mild assumptions. In particular, our solution is unique, matches conditional distributions across domains, recovers target proportions and explicitly controls the target generalization risk. Through an exhaustive comparison on several datasets, we challenge the state-of-the-art in GeTarS.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes