[hpc-announce] CFP: SI on "Reservoir Computing - Trends and Applications", Connection Science

[Editorial Manager]

* Apologies if multiple copies are received.  *


Special Issue on "Reservoir Computing - Trends and Applications"
Connection Science, Taylor & Francis
(SCI/SCIE JCR Q2 "best quartile", EI)
Web: http://shorturl.at/bltM0

Being inspired by the way in which the human brain processes information, the technique of artificial neural networks (ANN) is evolved to mimic human intelligence to solve complex problems. Thanks to its abilities, which has made its application successful across various fields of science and engineering (machine translation, pattern recognition, anomaly detection, decision making, etc.). Reservoir computing is an extensive model of ANN that deals with input signals connected to a fixed and random dynamical system. Such as creating a higher dimension representation called embedding, which is then connected to the desired output through trainable units. Here, the major role of the reservoir computing techniques is to transfer the sequential inputs non-linearly into a high-dimensional space. Thus the features of the inputs can be effectively interpreted using a simple learning algorithm. This paves the way for using faster learning algorithms to solve complex real-world problems.

The simplicity behind the training method in reservoir computing makes it application familiar across various streams. In practice, machine learning algorithms generally require higher computational power and larger training datasets. With the effective implementation of reservoir computing models, we can establish systems that can process the information faster with lesser learning cost and computational power. Recently, the features of reservoir computing have successfully extended across the domain of deep learning, leading to the development of innovative models such as deep reservoir computing and deep echo state networks. Further, reservoir computing systems are often used to train models that are used for the hierarchical processing of the temporal data and consequently enabling the exploration of the inherent potential of layered composition in recurrent neural network systems. Researchers worldwide are now speculating on the implications of reservoir computing across various disciplines. However, this remains still in infancy and requires greater improvements. In this regard, this special issue intends to find some of the recent trends in reservoir computing and its real-world applications in a more detailed manner. Researchers and practitioners working in this domain are most invited to present their novel and innovative research contributions.

Potential topics included, but not limited:
- Deep reservoir models and its applications
- Role of reservoir computing in neuroscience
- Echo state networks and reservoir computing
- Trends in reservoir computing for efficient big data analytics
- Novel reservoir computing models for emerging technologies
- Innovations in reservoir computing for machine learning and deep learning applications
- Trends in time-delay reservoir computing
- Novel and innovative applications of ensemble learning and reservoir computing
- Spiking dynamics and reservoir computing
- Training and learning inference in machine learning applications with reservoir computing models
- Reservoir computing and its influence on structured data

Manuscript deadline:  *15 February 2022*

Papers with technical contributions will be mainly considered, but survey papers may be considered only if of sufficient merit, and that strictly adhere to the theme of the special issue.

Special Issue Editor(s)
Dr. Qin Xin, Professor, Faculty of Science and Technology, University of the Faroe Islands, Faroe Islands, Denmark.
Dr. Avinab Marahatta, State Key Laboratory of Information Security, Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China
Dr. Dinesh Jackson Samuel, Faculty of Technology, Design and Environment, Oxford Brookes University, Oxford OX3 0BP, United Kingdom

For additional information, visit the webpage of this Special Issue or contact guest editor Dr. Qin Xin, email: qinx at ieee.org<mailto:qinx at ieee.org>