Advances in Randomized Neural Networks

Dear Colleagues,

In recent years, the study of Randomized Neural Networks has received a great deal of attention from the neural networks research community, making the randomized approach a successful, widely-popular, and well-attested paradigm for learning in a large class of domains. One of the major instances of this approach consists of the design of Neural Networks with Random Weights, where the connections to the hidden(s) layers are typically left untrained after initialization, and training is applied only to an output component (typically in the form of a linear learner). In the literature, this idea has been put forward and investigated in several forms, both in feed-forward architectures (e.g., under the names of random vector functional link, extreme learning machine, no-prop algorithm, and stochastic configuration network) and in recursive architectures (i.e., reservoir computing). Besides representing an easy-to-implement and extremely efficient design approach for learning machines, neural networks with random weights offer an illuminating perspective to the theoretical analysis of intrinsic architectural properties of neural systems. Recently, this paradigm has also been extended in the direction of deep neural networks, paving the way to a fruitful combination of the benefits of randomness and depth in the design of neural architectures. From a broader viewpoint, randomization can also enter the design of neural networks from other perspectives, including learning algorithms, regularization techniques, and ensemble methods.

This Special Issue calls for contributions that target the study and analysis of randomized neural networks from both theoretical and application viewpoints. The topics of interest include, but are not limited, to the following:

• neural networks with random weights
• randomized algorithms for neural networks
• random vector functional link, extreme learning machines, no-prop, and stochastic configuration networks
• reservoir computing, echo state networks, and liquid state machines
• deep neural networks with random weights (e.g., deep extreme learning machines and deep echo state networks)
• neural networks based on random projections
• physical implementations of randomized neural networks
• real-world applications

Dr. Claudio Gallicchio
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-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Informatics 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.