Learning Disentangled Semantic Representation for Domain Adaptation

Abstract

Domain adaptation is an important but challenging task. Most of the existing domain adaptation methods struggle to extract the domain-invariant representation on the feature space with entangling domain information and semantic information. Different from previous efforts on the entangled feature space, we aim to extract the domain invariant semantic information in the latent disentangled semantic representation (DSR) of the data. In DSR, we assume the data generation process is controlled by two independent sets of variables, i.e., the semantic latent variables and the domain latent variables. Under the above assumption, we employ a variational auto-encoder to reconstruct the semantic latent variables and domain latent variables behind the data. We further devise a dual adversarial network to disentangle these two sets of reconstructed latent variables. The disentangled semantic latent variables are finally adapted across the domains. Experimental studies testify that our model yields state-of-the-art performance on several domain adaptation benchmark datasets.

References in corpus (13)

• Auto-Encoding Variational Bayes
• Learning Transferable Features with Deep Adaptation Networks
• Unsupervised Domain Adaptation by Backpropagation
• Deep Domain Confusion: Maximizing for Domain Invariance
• Conditional Adversarial Domain Adaptation
• NICE: Non-linear Independent Components Estimation
• Unsupervised Domain Adaptation with Residual Transfer Networks
• Adversarial Discriminative Domain Adaptation
• Deep Transfer Learning with Joint Adaptation Networks
• Representation Learning: A Review and New Perspectives
• Progressive Feature Alignment for Unsupervised Domain Adaptation
• Domain-Invariant Adversarial Learning for Unsupervised Domain Adaption
• Domain Alignment with Triplets

Cited by in corpus (21)

• Learning Causal Semantic Representation for Out-of-Distribution Prediction
• An Unsupervised Domain Adaptation Model based on Dual-module Adversarial Training
• Continuous Domain Adaptation with Variational Domain-Agnostic Feature Replay
• Discriminative Radial Domain Adaptation
• Review-Based Domain Disentanglement without Duplicate Users or Contexts for Cross-Domain Recommendation
• Semi-Supervised Disentangled Framework for Transferable Named Entity Recognition
• Graph Domain Adaptation: Challenges, Progress and Prospects
• ToAlign: Task-oriented Alignment for Unsupervised Domain Adaptation
• Emo-DNA: Emotion Decoupling and Alignment Learning for Cross-Corpus Speech Emotion Recognition
• Learning to Match Distributions for Domain Adaptation
• Learning Disentangled Representations for Time Series
• Reducing the Covariate Shift by Mirror Samples in Cross Domain Alignment
• Towards Principled Disentanglement for Domain Generalization
• Review-Based Hyperbolic Cross-Domain Recommendation
• Multiple Classifiers Based Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
• Transferable Time-Series Forecasting under Causal Conditional Shift
• Transporting Causal Mechanisms for Unsupervised Domain Adaptation
• Disentanglement Then Reconstruction: Learning Compact Features for Unsupervised Domain Adaptation
• Bi-Directional Generation for Unsupervised Domain Adaptation
• Adversarial Dual Distinct Classifiers for Unsupervised Domain Adaptation
• Predicting Scientific Impact Through Diffusion, Conformity, and Contribution Disentanglement