Deep Unsupervised Domain Adaptation with Time Series Sensor Data: A Survey

Shi, Yongjie; Ying, Xianghua; Yang, Jinfa

doi:10.3390/s22155507

Cited by 23 publications

(7 citation statements)

References 150 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, we constructed an ANN model for the feature genes obtained from the above method according to the gene score by using the packages of “neuralnet” and “neuralnettools” in R software. The artificial neural network can simulate the structure and function of the brain neural network and deduce a set of classification rules from a set of disordered and irregular data, so as to realize the correct classification and construct a high-accuracy diagnosis model ( 38 ). Furthermore, the receiver operating characteristic (ROC) curve was utilized to evaluate the accuracy of the ANN model in the training and testing sets.…”

Section: Methodsmentioning

confidence: 99%

Deciphering the endometrial immune landscape of RIF during the window of implantation from cellular senescence by integrated bioinformatics analysis and machine learning

Zhao

Zhao²,

Jiang³

et al. 2022

Front. Immunol.

View full text Add to dashboard Cite

Recurrent implantation failure (RIF) is an extremely thorny issue in in-vitro fertilization (IVF)-embryo transfer (ET). However, its intricate etiology and pathological mechanisms are still unclear. Nowadays, there has been extensive interest in cellular senescence in RIF, and its involvement in endometrial immune characteristics during the window of implantation (WOI) has captured scholars’ growing concerns. Therefore, this study aims to probe into the pathological mechanism of RIF from cellular senescence and investigate the correlation between cellular senescence and endometrial immune characteristics during WOI based on bioinformatics combined with machine learning strategy, so as to elucidate the underlying pathological mechanisms of RIF and to explore novel treatment strategies for RIF. Firstly, the gene sets of GSE26787 and GSE111974 from the Gene Expression Omnibus (GEO) database were included for the weighted gene correlation network analysis (WGCNA), from which we concluded that the genes of the core module were closely related to cell fate decision and immune regulation. Subsequently, we identified 25 cellular senescence-associated differentially expressed genes (DEGs) in RIF by intersecting DEGs with cellular senescence-associated genes from the Cell Senescence (CellAge) database. Moreover, functional enrichment analysis was conducted to further reveal the specific molecular mechanisms by which these molecules regulate cellular senescence and immune pathways. Then, eight signature genes were determined by the machine learning method of support vector machine-recursive feature elimination (SVM-RFE), random forest (RF), and artificial neural network (ANN), comprising LATS1, EHF, DUSP16, ADCK5, PATZ1, DEK, MAP2K1, and ETS2, which were also validated in the testing gene set (GSE106602). Furthermore, distinct immune microenvironment abnormalities in the RIF endometrium during WOI were comprehensively explored and validated in GSE106602, including infiltrating immunocytes, immune function, and the expression profiling of human leukocyte antigen (HLA) genes and immune checkpoint genes. Moreover, the correlation between the eight signature genes with the endometrial immune landscape of RIF was also evaluated. After that, two distinct subtypes with significantly distinct immune infiltration characteristics were identified by consensus clustering analysis based on the eight signature genes. Finally, a “KEGG pathway–RIF signature genes–immune landscape” association network was constructed to intuitively uncover their connection. In conclusion, this study demonstrated that cellular senescence might play a pushing role in the pathological mechanism of RIF, which might be closely related to its impact on the immune microenvironment during the WOI phase. The exploration of the molecular mechanism of cellular senescence in RIF is expected to bring new breakthroughs for disease diagnosis and treatment strategies.

show abstract

Section: Methodsmentioning

confidence: 99%

Deciphering the endometrial immune landscape of RIF during the window of implantation from cellular senescence by integrated bioinformatics analysis and machine learning

Zhao

Zhao²,

Jiang³

et al. 2022

Front. Immunol.

View full text Add to dashboard Cite

show abstract

“…( 8)). Given that historical data of ⃗ r t is available for each REC and is also included in the input data, there is potential to extract cross-domain and domain-invariant features, a process known as domain adaptation [33][34][35].…”

Section: Time Series Processmentioning

confidence: 99%

Advancing Electric Load Forecasting: Leveraging Federated Learning for Distributed, Non-Stationary, and Discontinuous Time Series

Richter,

Lenk,

Bretschneider

2024

Preprint

View full text Add to dashboard Cite

This paper addresses the evolving landscape of electricity markets in Europe, with a focus on the integration of Renewable Energy Communities as introduced by the Renewable Energy Directive 2018/2001. Residents within a postal code area are highly incentivized to join a community, which enables them to exchange energy among themselves at lower procurement costs. Thereby, energy management systems optimize the operation of respective energy systems, with electrical load forecasting playing a key role. Given that prosumers may switch between communities on a daily basis, electricity demands within these groups will vary, leading to data that is non-stationary, discontinuous as well as non-identical and independently distributed. To encounter this issue, we propose a sophisticated forecasting model that applies federated learning, using informations from various distributed communities to learn domain-invariant features. To achieve this, we initially utilize synthetic electrical load time series at district level and aggregate them to profiles of Renewable Energy Communities with dynamic portfolios. Subsequently, we develop a forecasting model that accounts for the composition of residents of a Renewable Energy Community, adapt data pre-processing in accordance with the time series process, and detail a federated learning algorithm that incorporates weight averaging and data sharing. Following the training of various experimental setups, we ultimately evaluate their effectiveness. The findings suggest that our proposed framework is capable of effectively forecast non-stationary and discontinuous time series and that it can be applied to new, unseen data through the integration of knowledge from multiple sources.

show abstract

“…In most likelihood, it would be a subset of overall time-period representation that may impact the outcome. A survey on sensor time series [138] mentions that the strategies used for time-series DA bear much resemblance to non-time-series DA, with two specific strategies for time-series DAinput space adaptation and output space adaptation.…”

Section: ) Dependence On Different Time Period Subsetsmentioning

confidence: 99%

Domain Adaptation: Challenges, Methods, Datasets, and Applications

et al. 2023

View full text Add to dashboard Cite

Deep Neural Networks (DNNs) trained on one dataset (source domain) do not perform well on another set of data (target domain), which is different but has similar properties as the source domain. Domain Adaptation (DA) strives to alleviate this problem and has great potential in its application in practical settings, real-world scenarios, industrial applications and many data domains. Various DA methods aimed at individual data domains have been reported in the last few years; however, there is no comprehensive survey that encompasses all these data domains, focuses on the datasets available, the methods relevant to each domain, and importantly the applications and challenges. To that end, this survey paper discusses how DA can help DNNs work efficiently in these settings by reviewing DA methods and techniques. We have considered five data domains: computer vision, natural language processing, speech, time-series, and multi-modal data. We present a comprehensive taxonomy, including the methods, datasets, challenges, and applications corresponding to each domain. Our goal is to discuss industrial use cases and DA implementation for those. Our final aim is to provide future research directions based on evolving methods and results, the datasets used, and industrial applications.

show abstract

Deep Unsupervised Domain Adaptation with Time Series Sensor Data: A Survey

Cited by 23 publications

References 150 publications

Deciphering the endometrial immune landscape of RIF during the window of implantation from cellular senescence by integrated bioinformatics analysis and machine learning

Deciphering the endometrial immune landscape of RIF during the window of implantation from cellular senescence by integrated bioinformatics analysis and machine learning

Advancing Electric Load Forecasting: Leveraging Federated Learning for Distributed, Non-Stationary, and Discontinuous Time Series

Domain Adaptation: Challenges, Methods, Datasets, and Applications

Contact Info

Product

Resources

About