Adversarial Dropout Regularization

Abstract

We present a method for transferring neural representations from label-rich source domains to unlabeled target domains. Recent adversarial methods proposed for this task learn to align features across domains by fooling a special domain critic network. However, a drawback of this approach is that the critic simply labels the generated features as in-domain or not, without considering the boundaries between classes. This can lead to ambiguous features being generated near class boundaries, reducing target classification accuracy. We propose a novel approach, Adversarial Dropout Regularization (ADR), to encourage the generator to output more discriminative features for the target domain. Our key idea is to replace the critic with one that detects non-discriminative features, using dropout on the classifier network. The generator then learns to avoid these areas of the feature space and thus creates better features. We apply our ADR approach to the problem of unsupervised domain adaptation for image classification and semantic segmentation tasks, and demonstrate significant improvement over the state of the art. We also show that our approach can be used to train Generative Adversarial Networks for semi-supervised learning.

TBA on ICLR2018

References in corpus (8)

• Explaining and Harnessing Adversarial Examples
• Learning Transferable Features with Deep Adaptation Networks
• Unsupervised Domain Adaptation by Backpropagation
• Deep Domain Confusion: Maximizing for Domain Invariance
• FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation
• Adversarial Discriminative Domain Adaptation
• Unsupervised Cross-Domain Image Generation
• Deep CORAL: Correlation Alignment for Deep Domain Adaptation

Cited by in corpus (25)

• Universal Domain Adaptation through Self Supervision
• CDTrans: Cross-domain Transformer for Unsupervised Domain Adaptation
• Exploiting the Intrinsic Neighborhood Structure for Source-free Domain Adaptation
• Unsupervised Multi-Class Domain Adaptation: Theory, Algorithms, and Practice
• Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Data
• Should Graph Convolution Trust Neighbors? A Simple Causal Inference Method
• Self-ensembling for visual domain adaptation
• Survey of Dropout Methods for Deep Neural Networks
• Rethinking Distributional Matching Based Domain Adaptation
• FSDR: Frequency Space Domain Randomization for Domain Generalization
• Minimum Class Confusion for Versatile Domain Adaptation
• Contextual-Relation Consistent Domain Adaptation for Semantic Segmentation
• Learning Invariant Representations and Risks for Semi-supervised Domain Adaptation
• Semi-supervised Domain Adaptation based on Dual-level Domain Mixing for Semantic Segmentation
• Cross-Domain Gradient Discrepancy Minimization for Unsupervised Domain Adaptation
• Contrastively Smoothed Class Alignment for Unsupervised Domain Adaptation
• Curriculum based Dropout Discriminator for Domain Adaptation
• Semi-Supervised Domain Adaptation via Adaptive and Progressive Feature Alignment
• Learning Invariant Representation with Consistency and Diversity for Semi-supervised Source Hypothesis Transfer
• Discriminative Clustering for Robust Unsupervised Domain Adaptation
• Hypothesis Disparity Regularized Mutual Information Maximization
• Hard Class Rectification for Domain Adaptation
• Semi-Supervised Domain Adaptation via Selective Pseudo Labeling and Progressive Self-Training
• Exploring Dropout Discriminator for Domain Adaptation
• Adversarial Incremental Learning