New Trends in Machine Learning and Neural Computation

Dear Colleagues,

Machine Learning (ML) and Neural Computation have become fundamental components of modern informatics, providing computational methods for extracting knowledge from data, recognizing complex patterns, constructing predictive models, and supporting intelligent decision-making. Rapid developments in these fields are creating new opportunities for applications across science, engineering, economics, medicine, social sciences, and information systems.

The proposed Special Issue, “New Trends in Machine Learning and Neural Computation,” aims to provide a forum for recent theoretical advances, methodological developments, computational techniques, and emerging applications in machine learning and neural computation. Particular attention will be given to approaches that contribute to the development of reliable, interpretable, efficient, and mathematically well-founded computational models.

An important focus of the Special Issue is the development of new mathematical frameworks for machine learning and neural computation. In particular, we welcome research on learning methods based on hyper-complex calculus and hyper-complex-valued representations. Such approaches provide a natural framework for representing and processing multidimensional information and may lead to new learning architectures, optimization methods, regression techniques, and computational models.

The Special Issue also encourages research exploring connections between machine learning and other computational paradigms, including fuzzy systems, evolutionary computation, approximation theory, mathematical optimization, Bayesian methods, and computational intelligence.

Particular Areas of Interest

The Special Issue is especially interested in contributions that investigate new mathematical foundations and computational representations for machine learning. One such emerging direction is machine learning based on hyper-complex calculus, where hyper-complex algebra and differential calculus provide tools for representing multidimensional data and developing learning algorithms in a unified mathematical framework.

Potential contributions may address the formulation of learning algorithms in hyper-complex domains, hyper-complex differentiation and optimization, hyper-complex regression and approximation, hyper-complex neural networks, and applications involving multidimensional, vector-valued, or structured data. Another important direction concerns connections between machine learning and mathematically grounded computational methods, including fuzzy transforms, approximation methods, functional representations, Bayesian modeling, optimization, and numerical methods.

Such approaches may provide alternatives or complements to conventional neural-network architectures, particularly in situations where interpretability, approximation properties, limited data, uncertainty, multidimensional structure, or computational efficiency are important.

The Special Issue therefore welcomes both theoretical and application-oriented studies that establish relationships between different learning paradigms and investigate their mathematical foundations, computational properties, advantages, limitations, and areas of applicability.

Expected Contribution to the Field

The Special Issue is intended to bring together researchers working at the intersection of machine learning, neural computation, computational intelligence, mathematical analysis, and informatics. By combining theoretical perspectives with computational and application-oriented research, it aims to highlight emerging directions in the development of intelligent computational methods.

Particular emphasis will be placed on approaches that move beyond purely empirical model construction toward transparent, mathematically interpretable, robust, and efficient learning systems. The Special Issue will provide an opportunity to investigate how complementary mathematical and computational paradigms—including neural, probabilistic, fuzzy, approximation-based, and hyper-complex methods—can be integrated to address increasingly complex informatics problems.

A central objective is to encourage research in which new mathematical structures lead to new learning mechanisms, rather than being used merely as alternative representations of existing algorithms. This includes the development of new theories, algorithms, architectures, and applications based on hyper-complex calculus and related mathematical frameworks.

Prof. Dr. Irina Perfilieva
Guest Editor

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 250 words) can be sent to the Editorial Office for assessment.

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-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Information is an international peer-reviewed open access monthly 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 1800 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.