Visual Domain Adaptation with Manifold Embedded Distribution Alignment

TLDR

Visual domain adaptation seeks to learn robust target classifiers by leveraging source domain knowledge, yet existing methods struggle with degenerated feature transformations and inadequate evaluation of marginal versus conditional distribution alignment. This paper proposes Manifold Embedded Distribution Alignment (MEDA) to overcome these challenges. MEDA learns a domain‑invariant classifier on the Grassmann manifold via structural risk minimization while dynamically weighting marginal and conditional distributions for alignment. MEDA is the first to perform dynamic distribution alignment for manifold domain adaptation and achieves significant classification accuracy gains over state‑of‑the‑art traditional and deep methods.

Abstract

Visual domain adaptation aims to learn robust classifiers for the target domain by leveraging knowledge from a source domain. Existing methods either attempt to align the cross-domain distributions, or perform manifold subspace learning. However, there are two significant challenges: (1) degenerated feature transformation, which means that distribution alignment is often performed in the original feature space, where feature distortions are hard to overcome. On the other hand, subspace learning is not sufficient to reduce the distribution divergence. (2) unevaluated distribution alignment, which means that existing distribution alignment methods only align the marginal and conditional distributions with equal importance, while they fail to evaluate the different importance of these two distributions in real applications. In this paper, we propose a Manifold Embedded Distribution Alignment (MEDA) approach to address these challenges. MEDA learns a domain-invariant classifier in Grassmann manifold with structural risk minimization, while performing dynamic distribution alignment to quantitatively account for the relative importance of marginal and conditional distributions. To the best of our knowledge, MEDA is the first attempt to perform dynamic distribution alignment for manifold domain adaptation. Extensive experiments demonstrate that MEDA shows significant improvements in classification accuracy compared to state-of-the-art traditional and deep methods.