Quantum-Inspired Neural Networks

The rapid development in AI and quantum computing has resulted in a new domain termed Quantum-Inspired Neural Networks (QINNs). These models utilize ideas from quantum mechanics, including superposition, entanglement, and quantum probability, to improve the efficiency and performance of classical neural networks. Although full-scale quantum computing remains nascent, quantum-inspired methodologies have shown the capacity to address intricate computational challenges more effectively, even on conventional systems. In contrast to conventional neural networks that depend exclusively on classical optimization methods, QINNs integrate quantum-inspired strategies to improve learning velocity, generalization, and problem-solving capabilities.

This book seeks to examine the theoretical underpinnings, frameworks, and practical implementations of QINNs, rendering it an essential resource for scholars, academics, and industry experts. It examines mathematical frameworks behind quantum-inspired models, their implementation methodologies, and their relevance in diverse fields, including healthcare, finance, cybersecurity, and natural language processing. It serves as a comprehensive guide for individuals seeking to comprehend and apply QINNs in practical situations, utilizing theoretical insights, algorithmic frameworks, and case examples. The book is distinct due to its emphasis on the present and future of quantum-inspired deep learning. It integrates discussions on hybrid quantum-classical architectures, optimization strategies, and scalability difficulties, addressing the gap between quantum computing and classical AI, which are often treated separately in previous literature. Furthermore, it examines the constraints and future potential of QINNs, providing a framework for the shift from traditional deep learning to quantum-augmented models. Readers will acquire a profound comprehension of how quantum-inspired methodologies might transform the AI domain and propel innovation in nascent technologies.

Dr. Moolchand Sharma is an Assistant Professor at the Maharaja Agrasen Institute of Technology, GGSIPU Delhi. He has several publications in reputed international journals and conferences, including SCI-indexed and Scopus-indexed journals. He has authored/edited four books, and also has authored/co-authored chapters with in publications from reputed global publishers. His research areas include Artificial Intelligence, Nature-Inspired Computing, Security in Cloud Computing, Machine Learning, and Search Engine Optimization. He is associated with various professional bodies like IEEE, ISTE, IAENG, ICSES, UACEE, Internet Society, and a life member of the Universal Innovators research laboratory, etc. He possesses almost 10 years of teaching experience. He is the co-convener of the ICICC, DOSCI, ICDAM & ICCCN Springer Scopus-indexed conference series and ICCRDA-2020 Scopus-indexed Material Science & Engineering conference series. He is also the organizer and co-convener of the International Conference on Innovations and Ideas towards Patents (ICIIP) series. He is also the advisory and TPC committee member of the ICCIDS-2022 SSRN Conference. He is also the reviewer of many reputed journals. He has also served as a session chair in many international springer conferences. He has completed a PhD from DCR University of Science & Technology, Haryana. He completed his postgraduate studies in 2012 at SRM University, NCR/Ghaziabad, India and he graduated in 2010 from KNGD Modi Engineering College, Gautam Buddha Technical University.

Dr. Nebojsa Bacanin has a PhD from Faculty of Mathematics, University of Belgrade in 2015 (study program Computer Science, average grade 10,00). He was the vice-dean of the Graduate School of Computer Science and Faculty of Informatics and Computing in Belgrade, Serbia. He currently works as a Full Professor and as a Vice-Rector for Scientific Research at Singidunum University, Belgrade, Serbia. He is involved in scientific research in the field of computer science and his specialty includes artificial intelligence, machine learning, deep learning, stochastic optimization algorithms, swarm intelligence, soft-computing, optimization and modeling, image processing, computer vision and cloud and distributed computing. He actively works in the domain of novel and prospective research field, hybrid methods between machine learning and metaheuristics, where metaheuristics are applied for addressing non-deterministic polynomial hard (NP-hard) challenges from machine learning domain such as hyper-parameters optimization (tuning), training and feature selection. Besides improving machine learning/deep learning models for tackling various practical tasks for classification and regression, his research also involves optimized deep learning models for univariate and multivariate time-series forecasting. Moreover, he is an expert from the area of metaheuristics, and he has been actively doing research in enhancing swarm intelligence, as well as other types of metaheuristics, by incorporating minor changes (e.g., modification in exploitation/exploration expressions, parameters’ adjustments, etc.) and/or major modifications by performing hybridization with other methods (e.g., low-level and high-level hybrid metaheuristics methods). He has been applying his methods to wide variety of practical research areas, e.g., cloud computing scheduling, wireless sensor networks (WSNs) localization, coverage and energy consumption, X-ray images classification, stock price forecasting, portfolio optimization, as well as many others.

Dr. Tarik Ahmed Rashid is a Principal Fellow for the Higher Education Authority (PFHEA-UK) and a professor in the Department of Computer Science and Engineering at the University of Kurdistan Hewlêr, Iraq. He pursued his Post-Doctoral Fellowship at the Computer Science and Informatics School, College of Engineering, Mathematical and Physical Sciences, University College Dublin, Ireland. His research areas cover Artificial Intelligence, Nature Inspired Algorithms, Swarm Intelligence, Computational Intelligence, Machine Learning, and Data Mining. Tarik is among the Top 4 researchers in Iraq in the Web of Science-indexed published documents in engineering research filed over 5 years (2019-2023). He is on the prestigious Stanford University list of Top 2% of scientists in the world for 2021, 2022, 2023 and 2024. Tarik is also on the list of top 10 researchers in the Al-Ayen Iraqi Researchers Ranking (2022). AIR-Ranking 2022 is a national ranking organized by Al-Ayen University. His team has designed some single and multi-objective optimization algorithms, such as Fitness Dependent Optimizer (FDO), Child Drawing Development Optimization (CDDO), Donkey and smuggler optimization (DSO), Ant Nesting Algorithm (ANA), FOX Algorithm (FOX), Learner Performance based Behavior (LPB), Goose Algorithm (Goose), Lagrange Elementary Optimization (Leo), Shrike Optimization Algorithm (SHOA), Evolutionary Clustering Algorithm Star (ECA*), and Improved Evolutionary Clustering Algorithm Star (iECA*). His team also has designed several multi objective optimization algorithms, such as Multi-Objective Fitness Dependent Optimizer (MOFDO), Multi-objective Learner Performance based Behavior (MOLPB), and Multi-objective Ant Nesting Algorithm (ANA), and Grid Multi-objective Cat Swarm Optimization (GMOCSO).