scholarly journals A General Domain Specific Feature Transfer Framework for Hybrid Domain Adaptation

2019 ◽  
Vol 31 (8) ◽  
pp. 1440-1451 ◽  
Author(s):  
Pengfei Wei ◽  
Yiping Ke ◽  
Chi Keong Goh
Keyword(s):  
Domain Adaptation ◽  
General Domain ◽  
Domain Specific ◽  
Hybrid Domain

scholarly journals Domain Conditioned Adaptation Network

2020 ◽  
Vol 34 (07) ◽  
pp. 11386-11393 ◽  
Author(s):  
Shuang Li ◽  
Chi Liu ◽  
Qiuxia Lin ◽  
Binhui Xie ◽  
Zhengming Ding ◽  
...  
Keyword(s):  
Domain Adaptation ◽  
Feature Learning ◽  
Feature Representations ◽  
Domain Specific ◽  
Cross Domain ◽  
Learning Tasks ◽  
Domain Learning ◽  
High Level ◽  
High Level Feature ◽  
Target Data

Tremendous research efforts have been made to thrive deep domain adaptation (DA) by seeking domain-invariant features. Most existing deep DA models only focus on aligning feature representations of task-specific layers across domains while integrating a totally shared convolutional architecture for source and target. However, we argue that such strongly-shared convolutional layers might be harmful for domain-specific feature learning when source and target data distribution differs to a large extent. In this paper, we relax a shared-convnets assumption made by previous DA methods and propose a Domain Conditioned Adaptation Network (DCAN), which aims to excite distinct convolutional channels with a domain conditioned channel attention mechanism. As a result, the critical low-level domain-dependent knowledge could be explored appropriately. As far as we know, this is the first work to explore the domain-wise convolutional channel activation for deep DA networks. Moreover, to effectively align high-level feature distributions across two domains, we further deploy domain conditioned feature correction blocks after task-specific layers, which will explicitly correct the domain discrepancy. Extensive experiments on three cross-domain benchmarks demonstrate the proposed approach outperforms existing methods by a large margin, especially on very tough cross-domain learning tasks.

scholarly journals Word embeddings for application in geosciences: development, evaluation and examples of soil-related concepts

2019 ◽  
Author(s):  
José Padarian ◽  
Ignacio Fuentes
Keyword(s):  
Language Processing ◽  
Dimensional Space ◽  
Language Model ◽  
Test Suite ◽  
Word Embeddings ◽  
General Domain ◽  
Domain Specific ◽  
Descriptive Information ◽  
Development Evaluation ◽  
Numerical Representations

Abstract. A large amount of descriptive information is available in most disciplines of geosciences. This information is usually considered subjective and ill-favoured compared with its numerical counterpart. Considering the advances in natural language processing and machine learning, it is possible to utilise descriptive information and encode it as dense vectors. These word embeddings lay on a multi-dimensional space where angles and distances have a linguistic interpretation. We used 280 764 full-text scientific articles related to geosciences to train a domain-specific language model capable of generating such embeddings. To evaluate the quality of the numerical representations, we performed three intrinsic evaluations, namely: the capacity to generate analogies, term relatedness compared with the opinion of a human subject, and categorisation of different groups of words. Since this is the first attempt to evaluate word embedding for tasks in the geosciences domain, we created a test suite specific for geosciences. We compared our results with general domain embeddings commonly used in other disciplines. As expected, our domain-specific embeddings (GeoVec) outperformed general domain embeddings in all tasks, with an overall performance improvement of 107.9 %. The resulting embedding and test suite will be made available for other researchers to use an expand.

A Hybrid Domain Adaptation Approach for Identifying Crisis-Relevant Tweets

2019 ◽  
Vol 11 (2) ◽  
pp. 1-19 ◽  
Author(s):  
Reza Mazloom ◽  
Hongmin Li ◽  
Doina Caragea ◽  
Cornelia Caragea ◽  
Muhammad Imran
Keyword(s):  
Domain Adaptation ◽  
Supervised Machine Learning ◽  
Machine Learning Techniques ◽  
K Nearest Neighbors ◽  
Parameter Adaptation ◽  
Emergency Situations ◽  
Hybrid Domain ◽  
Learning Techniques ◽  
Source Data ◽  
Supervised Classifiers

Huge amounts of data generated on social media during emergency situations is regarded as a trove of critical information. The use of supervised machine learning techniques in the early stages of a crisis is challenged by the lack of labeled data for that event. Furthermore, supervised models trained on labeled data from a prior crisis may not produce accurate results, due to inherent crisis variations. To address these challenges, the authors propose a hybrid feature-instance-parameter adaptation approach based on matrix factorization, k-nearest neighbors, and self-training. The proposed feature-instance adaptation selects a subset of the source crisis data that is representative for the target crisis data. The selected labeled source data, together with unlabeled target data, are used to learn self-training domain adaptation classifiers for the target crisis. Experimental results have shown that overall the hybrid domain adaptation classifiers perform better than the supervised classifiers learned from the original source data.

scholarly journals Step by Step: Biology Undergraduates’ Problem-Solving Procedures during Multiple-Choice Assessment

2016 ◽  
Vol 15 (4) ◽  
pp. ar71 ◽  
Author(s):  
Luanna B. Prevost ◽  
Paula P. Lemons
Keyword(s):  
Problem Solving ◽  
Specific Problem ◽  
General Procedure ◽  
Multiple Choice ◽  
Higher Order ◽  
Hybrid Procedures ◽  
General Domain ◽  
Domain Specific ◽  
Written Descriptions ◽  
Categorization Scheme

This study uses the theoretical framework of domain-specific problem solving to explore the procedures students use to solve multiple-choice problems about biology concepts. We designed several multiple-choice problems and administered them on four exams. We trained students to produce written descriptions of how they solved the problem, and this allowed us to systematically investigate their problem-solving procedures. We identified a range of procedures and organized them as domain general, domain specific, or hybrid. We also identified domain-general and domain-specific errors made by students during problem solving. We found that students use domain-general and hybrid procedures more frequently when solving lower-order problems than higher-order problems, while they use domain-specific procedures more frequently when solving higher-order problems. Additionally, the more domain-specific procedures students used, the higher the likelihood that they would answer the problem correctly, up to five procedures. However, if students used just one domain-general procedure, they were as likely to answer the problem correctly as if they had used two to five domain-general procedures. Our findings provide a categorization scheme and framework for additional research on biology problem solving and suggest several important implications for researchers and instructors.

scholarly journals Unsupervised Domain Adaptation via Structured Prediction Based Selective Pseudo-Labeling

2020 ◽  
Vol 34 (04) ◽  
pp. 6243-6250 ◽  
Author(s):  
Qian Wang ◽  
Toby Breckon
Keyword(s):  
Domain Adaptation ◽  
Feature Space ◽  
Structured Prediction ◽  
Target Domain ◽  
Source Domain ◽  
Domain Specific ◽  
Unsupervised Domain Adaptation ◽  
Deep Feature ◽  
Significant Performance ◽  
Error Accumulation

Unsupervised domain adaptation aims to address the problem of classifying unlabeled samples from the target domain whilst labeled samples are only available from the source domain and the data distributions are different in these two domains. As a result, classifiers trained from labeled samples in the source domain suffer from significant performance drop when directly applied to the samples from the target domain. To address this issue, different approaches have been proposed to learn domain-invariant features or domain-specific classifiers. In either case, the lack of labeled samples in the target domain can be an issue which is usually overcome by pseudo-labeling. Inaccurate pseudo-labeling, however, could result in catastrophic error accumulation during learning. In this paper, we propose a novel selective pseudo-labeling strategy based on structured prediction. The idea of structured prediction is inspired by the fact that samples in the target domain are well clustered within the deep feature space so that unsupervised clustering analysis can be used to facilitate accurate pseudo-labeling. Experimental results on four datasets (i.e. Office-Caltech, Office31, ImageCLEF-DA and Office-Home) validate our approach outperforms contemporary state-of-the-art methods.

Sign in / Sign up

Export Citation Format

Share Document