Deep Transfer Network: Unsupervised Domain Adaptation

Abstract

Domain adaptation aims at training a classifier in one dataset and applying it to a related but not identical dataset. One successfully used framework of domain adaptation is to learn a transformation to match both the distribution of the features (marginal distribution), and the distribution of the labels given features (conditional distribution). In this paper, we propose a new domain adaptation framework named Deep Transfer Network (DTN), where the highly flexible deep neural networks are used to implement such a distribution matching process. This is achieved by two types of layers in DTN: the shared feature extraction layers which learn a shared feature subspace in which the marginal distributions of the source and the target samples are drawn close, and the discrimination layers which match conditional distributions by classifier transduction. We also show that DTN has a computation complexity linear to the number of training samples, making it suitable to large-scale problems. By combining the best paradigms in both worlds (deep neural networks in recognition, and matching marginal and conditional distributions in domain adaptation), we demonstrate by extensive experiments that DTN improves significantly over former methods in both execution time and classification accuracy.

References in corpus (1)

• Caffe: Convolutional Architecture for Fast Feature Embedding

Cited by in corpus (20)

• Deep Visual Domain Adaptation: A Survey
• Deep Transfer Learning for Person Re-identification
• A Comprehensive Survey of Deep Transfer Learning for Anomaly Detection in Industrial Time Series: Methods, Applications, and Directions
• Women also Snowboard: Overcoming Bias in Captioning Models
• Network Together: Node Classification via Cross network Deep Network Embedding
• Domain Adaptation with Conditional Distribution Matching and Generalized Label Shift
• Label Efficient Learning of Transferable Representations across Domains and Tasks
• Towards Fair Knowledge Transfer for Imbalanced Domain Adaptation
• A Novel Unsupervised Camera-aware Domain Adaptation Framework for Person Re-identification
• Iterative self-transfer learning: A general methodology for response time-history prediction based on small dataset
• Adversarial Alignment of Class Prediction Uncertainties for Domain Adaptation
• PIG: Prompt Images Guidance for Night-Time Scene Parsing
• Adversarial Camera Alignment Network for Unsupervised Cross-camera Person Re-identification
• Unsupervised Visual Domain Adaptation: A Deep Max-Margin Gaussian Process Approach
• Prototypical Cross-domain Knowledge Transfer for Cervical Dysplasia Visual Inspection
• Domain Adaptation with Randomized Expectation Maximization
• Multi-Source Domain Adaptation for Text Classification via DistanceNet-Bandits
• Exploring Bottom-up and Top-down Cues with Attentive Learning for Webly Supervised Object Detection
• Subdomain Adaptation with Manifolds Discrepancy Alignment
• Cross Domain Image Matching in Presence of Outliers