Minimal-Entropy Correlation Alignment for Unsupervised Deep Domain Adaptation

Abstract

In this work, we face the problem of unsupervised domain adaptation with a novel deep learning approach which leverages on our finding that entropy minimization is induced by the optimal alignment of second order statistics between source and target domains. We formally demonstrate this hypothesis and, aiming at achieving an optimal alignment in practical cases, we adopt a more principled strategy which, differently from the current Euclidean approaches, deploys alignment along geodesics. Our pipeline can be implemented by adding to the standard classification loss (on the labeled source domain), a source-to-target regularizer that is weighted in an unsupervised and data-driven fashion. We provide extensive experiments to assess the superiority of our framework on standard domain and modality adaptation benchmarks.

References in corpus (4)

• Adversarial Discriminative Domain Adaptation
• Unsupervised Cross-Domain Image Generation
• Associative Domain Adaptation
• A Compact Kernel Approximation for 3D Action Recognition

Cited by in corpus (6)

• BEV-Seg: Bird's Eye View Semantic Segmentation Using Geometry and Semantic Point Cloud
• A Prototype-Oriented Framework for Unsupervised Domain Adaptation
• SPG: Unsupervised Domain Adaptation for 3D Object Detection via Semantic Point Generation
• Automatic Validation of Textual Attribute Values in E-commerce Catalog by Learning with Limited Labeled Data
• Hypothesis Disparity Regularized Mutual Information Maximization
• Colorectal Polyp Classification from White-light Colonoscopy Images via Domain Alignment