Neuromorphic Computing for Computer Scientists

In 2019, Google astounded the world with the revelation that their quantum computer, Sycamore, had conquered an insurmountable problem. Remarkably, Sycamore achieved this feat in less than 200 seconds, a task that conventional computers, even the most potent ones, would require over 10,000 years to complete. While this achievement was truly impressive, one might ponder why it holds such immense significance. The answer lies in the profound implications of this breakthrough technology. Quantum computers have the potential to redefine the landscape of scientific discoveries, catalyze advancements in medicine, usher in a new era in artificial intelligence, and even play a pivotal role in averting catastrophic climate change. Yet, despite its remarkable promise, the widespread availability of quantum computers, in terms of qubit quantity, error-correction capabilities, and accessibility for practical applications, remains likely centuries away. Enter neuromorphic computing, a comparably groundbreaking alternative computing paradigm, accessible today and boasting similar efficiency, without the constraints associated with quantum computers. Neuromorphic circuits share fascinating similarities with quantum technologies, including the principles of superposition and quantum tunneling. This innovative approach is rapidly gaining momentum among researchers and leading global corporations, offering a powerful alternative. To put it into perspective, while Google's Sycamore registers a Q-Score of less than 140, neuromorphic platforms like the Dynex Cloud attain scores of 15,000 or more. In 'Neuromorphic Computing for Computer Scientists,' you will embark on an immersive journey into this cutting-edge field. This book is designed to be accessible yet rigorous, employing ideas and techniques familiar to any student of computer science, regardless of their mathematical or physics background. It begins with the fundamental prerequisites before delving into various facets of neuromorphic computing, viewed from a computer science perspective. Topics covered encompass computer architecture, algorithms, programming languages, theoretical computer science, cryptography, information theory, and hardware. The text offers practical examples, over two hundred exercises with solutions, and programming drills that breathe life into neuromorphic computing concepts. All code examples are runnable on the Dynex Neuromorphic Computing Platform, a cloud-based neuromorphic computing service. 'Neuromorphic Computing for Computer Scientists' invites readers to explore a realm of cutting-edge research, providing an engaging hands-on experience that simplifies intricate abstract concepts. This book is a valuable resource for computer science students and researchers, making the fascinating world of neuromorphic computing accessible to all. 1 Introduction 1 2 Quantum vs Dynex Neuromorphic Computing 3 3 Dynex: The Neuromorphic Cloud Computing Platform 15 4 Dynex SDK 18 5 Solving Problems with the Dynex SDK 23 6 Problem formulation 26 7 Model Classes 30 PART II: GETTING STARTED 8 Account Setup 34 9 Environment Setup & Installation 35 10 Testing the Setup 38 11 Hello, World! 40 PART III: PRACTICAL APPLICATIONS 12 Quantum Image Classification 43 13 4-Qubit Full Adder Circuit 52 14 Placement of Charging Stations 62 15 Quantum RNA Folding 70 16 Quantum Feature Selection 81 17 SciKit-Learn Plugin 91 18 Quantum-CFD 95 19 Quantum-SVM 99 20 Quantum-SISR 109 21 Quantum AI Security Auditing 116 22 Quantum Mode-Assisted AI Training 124 23 Quantum Prime Factorisation 137 25 Quantum Target Identification by Enzymes (TIE) 159 PART IV: DYNEX SDK REFERENCE 26 Dynex SDK Classes 168 27 Dynex SDK Methods 179 APPENDIX 189 GLOSSARY 228 ABOUT DYNEX 243