Mind the Class Weight Bias: Weighted Maximum Mean Discrepancy for Unsupervised Domain Adaptation

Abstract

In domain adaptation, maximum mean discrepancy (MMD) has been widely adopted as a discrepancy metric between the distributions of source and target domains. However, existing MMD-based domain adaptation methods generally ignore the changes of class prior distributions, i.e., class weight bias across domains. This remains an open problem but ubiquitous for domain adaptation, which can be caused by changes in sample selection criteria and application scenarios. We show that MMD cannot account for class weight bias and results in degraded domain adaptation performance. To address this issue, a weighted MMD model is proposed in this paper. Specifically, we introduce class-specific auxiliary weights into the original MMD for exploiting the class prior probability on source and target domains, whose challenge lies in the fact that the class label in target domain is unavailable. To account for it, our proposed weighted MMD model is defined by introducing an auxiliary weight for each class in the source domain, and a classification EM algorithm is suggested by alternating between assigning the pseudo-labels, estimating auxiliary weights and updating model parameters. Extensive experiments demonstrate the superiority of our weighted MMD over conventional MMD for domain adaptation.

10 pages, 5 figures, accepted by CVPR17

References in corpus (5)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Learning Transferable Features with Deep Adaptation Networks
• Unsupervised Domain Adaptation by Backpropagation
• Deep Domain Confusion: Maximizing for Domain Invariance

Cited by in corpus (21)

• Deep Visual Domain Adaptation: A Survey
• Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification
• Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-identification
• Deep Cocktail Network: Multi-source Unsupervised Domain Adaptation with Category Shift
• Learning to Adapt Invariance in Memory for Person Re-identification
• Racial Faces in-the-Wild: Reducing Racial Bias by Information Maximization Adaptation Network
• Open Set Domain Adaptation by Backpropagation
• Frustratingly Easy Person Re-Identification: Generalizing Person Re-ID in Practice
• Learning Factorized Representations for Open-set Domain Adaptation
• Domain Adaptation from Synthesis to Reality in Single-model Detector for Video Smoke Detection
• Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests
• Zero-Shot Deep Domain Adaptation
• Towards Discriminative Representation Learning for Unsupervised Person Re-identification
• Domain Adaptation and Image Classification via Deep Conditional Adaptation Network
• Unsupervised Domain Adaptation using Generative Models and Self-ensembling
• Hidden Covariate Shift: A Minimal Assumption For Domain Adaptation
• Heterogeneous Domain Adaptation via Soft Transfer Network
• Improving Unsupervised Domain Adaptation by Reducing Bi-level Feature Redundancy
• Adversarial Learning for Zero-shot Domain Adaptation
• Partially-Shared Variational Auto-encoders for Unsupervised Domain Adaptation with Target Shift
• Bi-Directional Generation for Unsupervised Domain Adaptation