A Brief Review of Domain Adaptation (2020-10-07T00:00:00.000000Z)

TL;DR

This paper focuses on unsupervised domain adaptation, where the labels are only available in the source domain, and presents some successful shallow and deep domain adaptation approaches that aim to deal with domain adaptation problems.

Abstract

Classical machine learning assumes that the training and test sets come from the same distributions. Therefore, a model learned from the labeled training data is expected to perform well on the test data. However, This assumption may not always hold in real-world applications where the training and the test data fall from different distributions, due to many factors, e.g., collecting the training and test sets from different sources, or having an out-dated training set due to the change of data over time. In this case, there would be a discrepancy across domain distributions, and naively applying the trained model on the new dataset may cause degradation in the performance. Domain adaptation is a sub-field within machine learning that aims to cope with these types of problems by aligning the disparity between domains such that the trained model can be generalized into the domain of interest. This paper focuses on unsupervised domain adaptation, where the labels are only available in the source domain. It addresses the categorization of domain adaptation from different viewpoints. Besides, It presents some successful shallow and deep domain adaptation approaches that aim to deal with domain adaptation problems.

Authors

Abolfazl Farahani

1 papers

Sahar Voghoei

1 papers

K. Rasheed

2 papers

TL;DR

Abstract

Authors

References84 items

The Bottleneck

DeepMSRF: A novel Deep Multimodal Speaker Recognition framework with Feature selection

DRDr: Automatic Masking of Exudates and Microaneurysms Caused By Diabetic Retinopathy Using Mask R-CNN and Transfer Learning

Video Contents Understanding using Deep Neural Networks

Personalized Feedback Emails: A Case Study on Online Introductory Computer Science Courses

On Parameter Tuning in Meta-Learning for Computer Vision

Universal Domain Adaptation

Contrastive Adaptation Network for Unsupervised Domain Adaptation

Deep Learning at the Edge

Dissection of Deep Learning with Applications in Image Recognition

Object Detection With Deep Learning: A Review

Multi-Adversarial Domain Adaptation

Open Set Domain Adaptation by Backpropagation

Importance Weighted Adversarial Nets for Partial Domain Adaptation

Deep learning for sentiment analysis: A survey

CyCADA: Cycle-Consistent Adversarial Domain Adaptation

Open Set Domain Adaptation

Unified Deep Supervised Domain Adaptation and Generalization

Squeeze-and-Excitation Networks

Partial Transfer Learning with Selective Adversarial Networks

Region based Image Steganalysis using Artificial Bee Colony

Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks

Adversarial Discriminative Domain Adaptation

Learning from Simulated and Unsupervised Images through Adversarial Training

Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks

Correlation Alignment for Unsupervised Domain Adaptation

Unsupervised Cross-Domain Image Generation

Multimodal Video Description

Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation

Coupled Generative Adversarial Networks

Deep Transfer Learning with Joint Adaptation Networks

Simultaneous Deep Transfer Across Domains and Tasks

A Neural Attention Model for Abstractive Sentence Summarization

Multimodal deep learning for robust RGB-D object recognition

Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models

U-Net: Convolutional Networks for Biomedical Image Segmentation

Deep Transfer Network: Unsupervised Domain Adaptation

Learning Transferable Features with Deep Adaptation Networks

DeepID-Net: Deformable deep convolutional neural networks for object detection

Domain-Adversarial Neural Networks

Learning Spatiotemporal Features with 3D Convolutional Networks

Unsupervised Domain Adaptation by Backpropagation

Collaborative Deep Learning for Recommender Systems

Generative Adversarial Nets

Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation

Generalized Transfer Subspace Learning Through Low-Rank Constraint

Transfer Feature Learning with Joint Distribution Adaptation

Unsupervised Visual Domain Adaptation Using Subspace Alignment

Domain Generalization via Invariant Feature Representation

Foundations of Machine Learning

Marginalized Denoising Autoencoders for Domain Adaptation

Robust visual domain adaptation with low-rank reconstruction

Geodesic flow kernel for unsupervised domain adaptation

Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition

Domain adaptation for object recognition: An unsupervised approach

Collaborative topic modeling for recommending scientific articles

Domain Adaptation for Large-Scale Sentiment Classification: A Deep Learning Approach

A Survey on Transfer Learning

Binary coding of speech spectrograms using a deep auto-encoder

Deconvolutional networks

A theory of learning from different domains

Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion

Multi-view learning of acoustic features for speaker recognition

Domain Adaptation via Transfer Component Analysis

Covariate Shift by Kernel Mean Matching

A Survey of Semi-Supervised Learning Methods

Direct Importance Estimation with Model Selection and Its Application to Covariate Shift Adaptation

Covariate Shift Adaptation by Importance Weighted Cross Validation

Correcting Sample Selection Bias by Unlabeled Data

Analysis of Representations for Domain Adaptation

A Kernel Method for the Two-Sample-Problem

Model Selection Under Covariate Shift

Input-dependent estimation of generalization error under covariate shift

Learning and evaluating classifiers under sample selection bias

Nonparametric and Semiparametric Models

Improving predictive inference under covariate shift by weighting the log-likelihood function