Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Data

Abstract

Unsupervised domain adaptation aims to align a labeled source domain and an unlabeled target domain, but it requires to access the source data which often raises concerns in data privacy, data portability and data transmission efficiency. We study unsupervised model adaptation (UMA), or called Unsupervised Domain Adaptation without Source Data, an alternative setting that aims to adapt source-trained models towards target distributions without accessing source data. To this end, we design an innovative historical contrastive learning (HCL) technique that exploits historical source hypothesis to make up for the absence of source data in UMA. HCL addresses the UMA challenge from two perspectives. First, it introduces historical contrastive instance discrimination (HCID) that learns from target samples by contrasting their embeddings which are generated by the currently adapted model and the historical models. With the historical models, HCID encourages UMA to learn instance-discriminative target representations while preserving the source hypothesis. Second, it introduces historical contrastive category discrimination (HCCD) that pseudo-labels target samples to learn category-discriminative target representations. Specifically, HCCD re-weights pseudo labels according to their prediction consistency across the current and historical models. Extensive experiments show that HCL outperforms and state-of-the-art methods consistently across a variety of visual tasks and setups.

arXiv commentAccepted to Advances in Neural Information Processing Systems 34 (NeurIPS 2021)

References in corpus (22)

• Learning Transferable Features with Deep Adaptation Networks
• Learning deep representations by mutual information estimation and maximization
• Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
• Temporal Ensembling for Semi-Supervised Learning
• ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation
• Semantic Foggy Scene Understanding with Synthetic Data
• Unsupervised Domain Adaptation with Residual Transfer Networks
• Curriculum Domain Adaptation for Semantic Segmentation of Urban Scenes
• Contrastive Multiview Coding
• Model Adaptation: Unsupervised Domain Adaptation without Source Data
• Self-ensembling for visual domain adaptation
• Adversarial Dropout Regularization
• Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervision
• MLAN: Multi-Level Adversarial Network for Domain Adaptive Semantic Segmentation
• Source-Free Domain Adaptation for Semantic Segmentation
• A Free Lunch for Unsupervised Domain Adaptive Object Detection without Source Data
• Category Contrast for Unsupervised Domain Adaptation in Visual Tasks
• Co${}^2$L: Contrastive Continual Learning
• Semi-Supervised Domain Adaptation via Adaptive and Progressive Feature Alignment
• Spectral Unsupervised Domain Adaptation for Visual Recognition
• RDA: Robust Domain Adaptation via Fourier Adversarial Attacking
• Domain Adaptive Video Segmentation via Temporal Consistency Regularization

Cited by in corpus (4)

• A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts
• Cal-SFDA: Source-Free Domain-adaptive Semantic Segmentation with Differentiable Expected Calibration Error
• Source-Free Domain Adaptation for Question Answering with Masked Self-training
• Energy-Based Pseudo-Label Refining for Source-free Domain Adaptation