Neural Architecture Search

Dear Colleagues,

In recent years, Deep Learning has quickly been becoming a de facto standard for solving real world problems of very diverse kinds. Techniques such as convolutional neural networks are outstanding performers when tackling computer vision problems, LSTMs and other recurrent architectures are proficiently solving natural language processing and understanding problems, and deep learning is in general being considered as a promising approach for many other domains, including games or medicine.

However, a problem remains inherent to the use of deep learning techniques: Designing a functional architecture remains not an easy task. The process generally involves large amounts of trial-and-error to find suitable architectures that attain a reasonable performance in the problem of choice. In this context, neural architecture search unveils as a useful alternative for automatically finding good-performing topologies. Further, the search procedure can be extended to also find optimal hyperparameters for the learning process.

This Special Issue on Neural Architecture Search calls for manuscripts describing innovations and novel applications of neural architecture search. We invite researchers to contribute original research papers devoted to advance in this field. Topics relevant to the Special Issue include but are not limited to the following:

• Neuroevolution of deep learning architectures;
• Novel search methods for architecture optimization;
• Self-tuning of learning parameters;
• Automatic discovery of novel deep learning architectures;
• Multiobjective optimization in deep learning networks;
• Automatic performance optimization in deep learning;
• Evolution or optimization of adversarial models;
• Collaborative or competitive evolution in deep learning;
• Applications of neural architecture search.

Dr. Alejandro Baldominos
Dr. Alejandro Martín
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. AI is an international peer-reviewed open access quarterly 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 1600 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.

• Neural architecture search
• Deep learning
• Neural networks
• Neuroevolution
• Artificial intelligence