A review of domain adaptation without target labels.

08:00 EDT 7th October 2019 | BioPortfolio

Summary of "A review of domain adaptation without target labels."

Domain adaptation has become a prominent problem setting in machine learning and related fields. This review asks the question: how can a classifier learn from a source domain and generalize to a target domain? We present a categorization of approaches, divided into, what we refer to as, sample-based, feature-based and inference-based methods. Sample-based methods focus on weighting individual observations during training based on their importance to the target domain. Feature-based methods revolve around on mapping, projecting and representing features such that a source classifier performs well on the target domain and inference-based methods incorporate adaptation into the parameter estimation procedure, for instance through constraints on the optimization procedure. Additionally, we review a number of conditions that allow for formulating bounds on the cross-domain generalization error. Our categorization highlights recurring ideas and raises questions important to further research.


Journal Details

This article was published in the following journal.

Name: IEEE transactions on pattern analysis and machine intelligence
ISSN: 1939-3539


DeepDyve research library

PubMed Articles [20118 Associated PubMed Articles listed on BioPortfolio]

Inferring Latent Domains for Unsupervised Deep Domain Adaptation.

Unsupervised Domain Adaptation (UDA) refers to the problem of learning a model in a target domain where labeled data are not available by leveraging information from annotated data in a source domain....

Multiclass heterogeneous domain adaptation via bidirectional ECOC projection.

Heterogeneous domain adaptation aims to exploit the source domain data to train a prediction model for the target domain with different input feature space. Current methods either map the data points ...

Unsupervised Domain Adaptation for Depth Prediction from Images.

State-of-the-art methods to infer dense and accurate depth measurements from images rely on deep CNN models trained in an end-to-end fashion on a significant amount of data. However, despite the outst...

Multi-representation adaptation network for cross-domain image classification.

In image classification, it is often expensive and time-consuming to acquire sufficient labels. To solve this problem, domain adaptation often provides an attractive option given a large amount of lab...

Unsupervised Domain Adaptation With Adversarial Residual Transform Networks.

Domain adaptation (DA) is widely used in learning problems lacking labels. Recent studies show that deep adversarial DA models can make markable improvements in performance, which include symmetric an...

Clinical Trials [4129 Associated Clinical Trials listed on BioPortfolio]

The Effect of Menu Labels on Calories Ordered in Hispanics

Most studies on menu labels have been conducted in mostly non-Hispanic white individuals, and how menu labels affect calories and macronutrients ordered in Hispanics is not clearly underst...

Testing the Influence of Different Sugary Drink Warning Label Designs

The aim of this study is to compare different sugary drink labels to see which most influence knowledge, beliefs, purchase intentions and choices. Parents of children ages 6-11 will shop i...

A Randomized-controlled Experiment of Sugary Drink Warning Labels

The primary objective of this study is to determine to what degree sugary drink warning labels increase consumers' knowledge about the potential health harms of sugary drinks and reduce su...

Improving Prescription Drug Warning Labels

Prior studies have documented a high prevalence of patients misunderstanding prescription drug warning labels, placing them at risk for medication error. We sought to evaluate the effecti...

MRI Study of Saccadic Adaptation

Accurate execution of our directed behaviors in the immediate environment allows the vast majority of our daily activities. Through the movements of our eyes and different parts of our bod...

Medical and Biotech [MESH] Definitions

Review of the medical necessity of hospital or other health facility admissions, upon or within a short time following an admission, and periodic review of services provided during the course of treatment.

The adjustment of the eye to variations in the intensity of light. Light adaptation is the adjustment of the eye when the light threshold is increased; DARK ADAPTATION when the light is greatly reduced. (From Cline et al., Dictionary of Visual Science, 4th ed)

Formal programs for assessing drug prescription against some standard. Drug utilization review may consider clinical appropriateness, cost effectiveness, and, in some cases, outcomes. Review is usually retrospective, but some analysis may be done before drugs are dispensed (as in computer systems which advise physicians when prescriptions are entered). Drug utilization review is mandated for Medicaid programs beginning in 1993.

Organizations representing designated geographic areas which have contracts under the PRO program to review the medical necessity, appropriateness, quality, and cost-effectiveness of care received by Medicare beneficiaries. Peer Review Improvement Act, PL 97-248, 1982.

The caring for individuals in institutions and their adaptation to routines characteristic of the institutional environment, and/or their loss of adaptation to life outside the institution.

Quick Search

DeepDyve research library

Searches Linking to this Article