Neural Networks for Feature Extraction

A special issue of Electronics (ISSN 2079-9292). This special issue belongs to the section "Computer Science & Engineering".

Deadline for manuscript submissions: 31 March 2024 | Viewed by 171

Special Issue Editors

School of Artificial Intelligence, Xidian University, Xi’an 710071, China
Interests: artificial intelligence
Academy of Advanced Disciplinary Research, Xidian University, Xi’an 710071, China
Interests: artificial intelligence; medical imaging; machine Learning

Special Issue Information

Dear Colleagues,

Neural networks are powerful machine learning algorithms that can automatically learn feature representations of input data. Compared with traditional methods, using neural networks for feature extraction has the following advantages: as an unsupervised feature learning method, neural networks can automatically discover features in the dataset without manual feature engineering. Deep neural networks can learn more abstract and advanced features of the data, containing information about the attributes and structure of the data, while the features learned by traditional methods tend to be more-superficial and harder-to-learn, abstract features. The features learned by neural networks can be well generalized to new data, while the features learned by traditional methods tend to be too dependent on training data. Neural networks can learn the intrinsic dependencies between features, and such feature representations are often more powerful than simply stacking features.

Neural networks have been widely used in feature extraction tasks, and current research focuses on the following aspects: improving the effectiveness of feature learning, e.g., through regularization and pre-training; and designing new neural network structures to learn more abstract and efficient features, e.g., convolutional neural networks, recurrent neural networks, and spike neural networks.
Combining neural networks with other methods can be carried out to form a more powerful feature learning framework.

In summary, neural networks have the advantages of automatically learning features, learning abstract features, good feature generalization, and convergence to stable features, which often make the features learned by neural networks more powerful than artificial features and widely applicable to downstream tasks. Neural networks provide a powerful tool for feature learning and representation.

Dr. Zhen Cao
Dr. Zhang Guo
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers

This special issue is now open for submission.
Back to TopTop