Beyond Sharing Weights for Deep Domain Adaptation

Abstract

The performance of a classifier trained on data coming from a specific domain typically degrades when applied to a related but different one. While annotating many samples from the new domain would address this issue, it is often too expensive or impractical. Domain Adaptation has therefore emerged as a solution to this problem; It leverages annotated data from a source domain, in which it is abundant, to train a classifier to operate in a target domain, in which it is either sparse or even lacking altogether. In this context, the recent trend consists of learning deep architectures whose weights are shared for both domains, which essentially amounts to learning domain invariant features. Here, we show that it is more effective to explicitly model the shift from one domain to the other. To this end, we introduce a two-stream architecture, where one operates in the source domain and the other in the target domain. In contrast to other approaches, the weights in corresponding layers are related but not shared. We demonstrate that this both yields higher accuracy than state-of-the-art methods on several object recognition and detection tasks and consistently outperforms networks with shared weights in both supervised and unsupervised settings.

References in corpus (9)

• ADADELTA: An Adaptive Learning Rate Method
• Learning Transferable Features with Deep Adaptation Networks
• Deep Domain Confusion: Maximizing for Domain Invariance
• Unsupervised Domain Adaptation with Residual Transfer Networks
• Progressive Neural Networks
• Domain Separation Networks
• Domain Generalization via Invariant Feature Representation
• Memory Bounded Deep Convolutional Networks
• On Rendering Synthetic Images for Training an Object Detector

Cited by in corpus (86)

• Deep Visual Domain Adaptation: A Survey
• Model Adaptation: Unsupervised Domain Adaptation without Source Data
• Adversarial Discriminative Domain Adaptation
• Domain Adaptation for Visual Applications: A Comprehensive Survey
• Few-Shot Adversarial Domain Adaptation
• Heterogeneous domain adaptation: An unsupervised approach
• Unified Image and Video Saliency Modeling
• Unsupervised Domain Adaptation with Adversarial Residual Transform Networks
• Unsupervised Multi-Class Domain Adaptation: Theory, Algorithms, and Practice
• Coupled Generative Adversarial Networks
• Domain Adaptation by Class Centroid Matching and Local Manifold Self-Learning
• Domain Adaptation on Point Clouds via Geometry-Aware Implicits
• Label Efficient Learning of Transferable Representations across Domains and Tasks
• Adversarial adaptive 1-D convolutional neural networks for bearing fault diagnosis under varying working condition
• Dynamic Instance Domain Adaptation
• A Sentinel-2 multi-year, multi-country benchmark dataset for crop classification and segmentation with deep learning
• FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare
• Associative Domain Adaptation
• Meta-FDMixup: Cross-Domain Few-Shot Learning Guided by Labeled Target Data
• Adversarial-Prediction Guided Multi-task Adaptation for Semantic Segmentation of Electron Microscopy Images
• An Annotation Saved is an Annotation Earned: Using Fully Synthetic Training for Object Instance Detection
• M3BAT: Unsupervised Domain Adaptation for Multimodal Mobile Sensing with Multi-Branch Adversarial Training
• Joint Domain Alignment and Discriminative Feature Learning for Unsupervised Deep Domain Adaptation
• Robust Local Preserving and Global Aligning Network for Adversarial Domain Adaptation
• Domain Adaptation for Semantic Segmentation via Patch-Wise Contrastive Learning
• A Review of Single-Source Deep Unsupervised Visual Domain Adaptation
• Unsupervised Neural Domain Adaptation for Document Image Binarization
• SALIENCE: An Unsupervised User Adaptation Model for Multiple Wearable Sensors Based Human Activity Recognition
• Active Long Term Memory Networks
• See Eye to Eye: A Lidar-Agnostic 3D Detection Framework for Unsupervised Multi-Target Domain Adaptation
• Vicinal and categorical domain adaptation
• Learning Vision-based Cohesive Flight in Drone Swarms
• Correlation Alignment by Riemannian Metric for Domain Adaptation
• Anatomy-guided domain adaptation for 3D in-bed human pose estimation
• Domain-Invariant Adversarial Learning for Unsupervised Domain Adaption
• Attention Regularized Laplace Graph for Domain Adaptation
• A Comprehensive Analysis of Information Leakage in Deep Transfer Learning
• Learning Factorized Representations for Open-set Domain Adaptation
• Online Adaptation to Label Distribution Shift
• A Multi Camera Unsupervised Domain Adaptation Pipeline for Object Detection in Cultural Sites through Adversarial Learning and Self-Training
• Domain Adaptation with Auxiliary Target Domain-Oriented Classifier
• Deep Semantic Parsing of Freehand Sketches with Homogeneous Transformation, Soft-Weighted Loss, and Staged Learning
• Deep Adversarial Transition Learning using Cross-Grafted Generative Stacks
• Curriculum based Dropout Discriminator for Domain Adaptation
• Multimodal Multi-User Surface Recognition with the Kernel Two-Sample Test
• Emo-DNA: Emotion Decoupling and Alignment Learning for Cross-Corpus Speech Emotion Recognition
• Domain-Adaptive Few-Shot Learning
• Fast CRDNN: Towards on Site Training of Mobile Construction Machines
• A microservice-based framework for exploring data selection in cross-building knowledge transfer
• AED-PADA:Improving Generalizability of Adversarial Example Detection via Principal Adversarial Domain Adaptation
• Multi-Stage Transfer Learning with an Application to Selection Process
• Toward Learning Human-aligned Cross-domain Robust Models by Countering Misaligned Features
• Factorized Adversarial Networks for Unsupervised Domain Adaptation
• Unsupervised Domain Adaptation: from Simulation Engine to the RealWorld
• Federated Multi-Target Domain Adaptation
• Lifelong Learning with Sketched Structural Regularization
• Dynamic Feature Alignment for Semi-supervised Domain Adaptation
• Towards Self-similarity Consistency and Feature Discrimination for Unsupervised Domain Adaptation
• Locality Preserving Joint Transfer for Domain Adaptation
• Unsupervised Domain Adaptation for Learning Eye Gaze from a Million Synthetic Images: An Adversarial Approach
• Unsupervised Domain Adaptation with Temporal-Consistent Self-Training for 3D Hand-Object Joint Reconstruction
• Multi-Source Domain Adaptation for Text Classification via DistanceNet-Bandits
• Tackling Imbalanced Data in Cybersecurity with Transfer Learning: A Case with ROP Payload Detection
• Unsupervised Domain Adaptation using Deep Networks with Cross-Grafted Stacks
• Joint Learning of Neural Networks via Iterative Reweighted Least Squares
• Introducing Pose Consistency and Warp-Alignment for Self-Supervised 6D Object Pose Estimation in Color Images
• Discriminative Noise Robust Sparse Orthogonal Label Regression-based Domain Adaptation
• Adversarial Feature Training for Generalizable Robotic Visuomotor Control
• A Dictionary Approach to Domain-Invariant Learning in Deep Networks
• Enlarging Discriminative Power by Adding an Extra Class in Unsupervised Domain Adaptation
• Unsupervised Sentiment Analysis by Transferring Multi-source Knowledge
• Towards Robust Pattern Recognition: A Review
• Multi-Purposing Domain Adaptation Discriminators for Pseudo Labeling Confidence
• Sim2Real for Self-Supervised Monocular Depth and Segmentation
• Interventional Domain Adaptation
• Geometry-Aware Self-Training for Unsupervised Domain Adaptationon Object Point Clouds
• Exploring Dropout Discriminator for Domain Adaptation
• FDA: Feature Decomposition and Aggregation for Robust Airway Segmentation
• Domain Adaptation and Active Learning for Fine-Grained Recognition in the Field of Biodiversity
• Towards Uncovering the Intrinsic Data Structures for Unsupervised Domain Adaptation using Structurally Regularized Deep Clustering
• CHEER: Rich Model Helps Poor Model via Knowledge Infusion
• A Simple Domain Shifting Networkfor Generating Low Quality Images
• Learning to Align Images using Weak Geometric Supervision
• Autoencoder Based Sample Selection for Self-Taught Learning
• Unsupervised Domain Adaptive Object Detection using Forward-Backward Cyclic Adaptation
• DAVOS: Semi-Supervised Video Object Segmentation via Adversarial Domain Adaptation