Photonic Artificial Intelligence

Neuromorphic Photonic Devices and Applications synthesizes the most critical advances in photonic neuromorphic models, photonic material platforms and accelerators for neuromorphic computing. The book discusses fields and applications that can leverage these new platforms. A brief review of the historical development of the field is followed by a discussion of the emerging 2D photonic materials platforms and recent work in implementing neuromorphic models and 3D neuromorphic systems. The application of artificial intelligence (AI), such as neuromorphic models to inverse design neuromorphic materials and devices and predict performance challenges is discussed throughout. Finally, a comprehensive overview of the applications of neuromorphic photonic technologies and the challenges, opportunities and future prospects is discussed, making the book suitable for researchers and practitioners in academia and R&D in the multidisciplinary field of photonics. Includes overview of primary scientific concepts for the research topic of neuromorphic photonics such as neurons as computational units, artificial intelligence, machine learning and neuromorphic models Reviews the latest advances in photonic materials, device platforms and enabling technology drivers of neuromorphic photonics Discusses potential applications in computing and optical communications

This book sets out to build bridges between the domains of photonic device physics and neural networks, providing a comprehensive overview of the emerging field of "neuromorphic photonics." It includes a thorough discussion of evolution of neuromorphic photonics from the advent of fiber-optic neurons to today’s state-of-the-art integrated laser neurons, which are a current focus of international research. Neuromorphic Photonics explores candidate interconnection architectures and devices for integrated neuromorphic networks, along with key functionality such as learning. It is written at a level accessible to graduate students, while also intending to serve as a comprehensive reference for experts in the field.

Artificial intelligence is deeply involved in our daily lives via reinforcing the digital transformation of modern economies and infrastructure. It relies on powerful computing clusters, which face bottlenecks of power consumption for both data transmission and intensive computing. Meanwhile, optics (especially optical communications, which underpin today’s telecommunications) is penetrating short-reach connections down to the chip level, thus meeting with AI technology and creating numerous opportunities. This book is about the marriage of optics and AI and how each part can benefit from the other. Optics facilitates on-chip neural networks based on fast optical computing and energy-efficient interconnects and communications. On the other hand, AI enables efficient tools to address the challenges of today’s optical communication networks, which behave in an increasingly complex manner. The book collects contributions from pioneering researchers from both academy and industry to discuss the challenges and solutions in each of the respective fields.

This book, the first of its kind, bridges the gap between the increasingly interlinked fields of nanophotonics and artificial intelligence (AI). While artificial intelligence techniques, machine learning in particular, have revolutionized many different areas of scientific research, nanophotonics holds a special position as it simultaneously benefits from AI-assisted device design whilst providing novel computing platforms for AI. This book is aimed at both researchers in nanophotonics who want to utilize AI techniques and researchers in the computing community in search of new photonics-based hardware. The book guides the reader through the general concepts and specific topics of relevance from both nanophotonics and AI, including optical antennas, metamaterials, metasurfaces, and other photonic devices on the one hand, and different machine learning paradigms and deep learning algorithms on the other. It goes on to comprehensively survey inverse techniques for device design, AI-enabled applications in nanophotonics, and nanophotonic platforms for AI. This book will be essential reading for graduate students, academic researchers, and industry professionals from either side of this fast-developing, interdisciplinary field.

This book delves into optics and photonic materials, describing the development of an intelligent all-optical system capable of replicating the functional building blocks of the biological brain. Starting with an analysis of biological neuronal dynamics and traversing the state of the art of neuromorphic systems developed to date, the book arrives at a description of neural networks realized through spatial soliton technology. After a brief introduction to the biology of neural networks (Chapter 1), the book delves into the description of the neuromorphic problem emphasizing the peculiarities of optical hardware developed to date. (Chapter 2). Chapter 3 is dedicated to the description of psychomemories , which represent the modeling of human learning according to the theories of modern neuro-psychology. This chapter provides the prerequisites for understanding how solitonic neural networks (SNNs) are able to learn and how they approach biological models. Chapter 4 focuses on the experimentation of solitonic optic neurons in thin layers of lithium niobate. Optical techniques for supervised and unsupervised learning are discussed. The entire chapter is accompanied by theoretical, simulative and experimental results. This chapter explains how an X-junction neuron is able to establish synapses, modify them, or erase them. The erasure of solitonic structures represents an important innovation in the field of nonlinear optics. Finally, Chapter 5 reports on the implementation of a network of neurons capable of processing information and storing it exactly as a human episodic memory does. The chapter ends with a number of insights into the lines of research that are currently being pursued on the basis of the results obtained. The book is meant for graduate students and researchers in the fields of optics, photonic applications, and biology. However, the main beneficiaries of this book are senior researchers in the field of nonlinear optics and artificial intelligence. To fully understand the results, it is important to have a basic knowledge of optical physics and neuron biology.

Silicon photonics is beginning to play an important role in driving innovations in communication and computation for an increasing number of applications, from health care and biomedical sensors to autonomous driving, datacenter networking, and security. In recent years, there has been a significant amount of effort in industry and academia to innovate, design, develop, analyze, optimize, and fabricate systems employing silicon photonics, shaping the future of not only Datacom and telecom technology but also high-performance computing and emerging computing paradigms, such as optical computing and artificial intelligence. Different from existing books in this area, Silicon Photonics for High-Performance Computing and Beyond presents a comprehensive overview of the current state-of-the-art technology and research achievements in applying silicon photonics for communication and computation. It focuses on various design, development, and integration challenges, reviews the latest advances spanning materials, devices, circuits, systems, and applications. Technical topics discussed in the book include: • Requirements and the latest advances in high-performance computing systems • Device- and system-level challenges and latest improvements to deploy silicon photonics in computing systems • Novel design solutions and design automation techniques for silicon photonic integrated circuits • Novel materials, devices, and photonic integrated circuits on silicon • Emerging computing technologies and applications based on silicon photonics Silicon Photonics for High-Performance Computing and Beyond presents a compilation of 19 outstanding contributions from academic and industry pioneers in the field. The selected contributions present insightful discussions and innovative approaches to understand current and future bottlenecks in high-performance computing systems and traditional computing platforms, and the promise of silicon photonics to address those challenges. It is ideal for researchers and engineers working in the photonics, electrical, and computer engineering industries as well as academic researchers and graduate students (M.S. and Ph.D.) in computer science and engineering, electronic and electrical engineering, applied physics, photonics, and optics.

Machine Learning for Future Fiber-Optic Communication Systems provides a comprehensive and in-depth treatment of machine learning concepts and techniques applied to key areas within optical communications and networking, reflecting the state-of-the-art research and industrial practices. The book gives knowledge and insights into the role machine learning-based mechanisms will soon play in the future realization of intelligent optical network infrastructures that can manage and monitor themselves, diagnose and resolve problems, and provide intelligent and efficient services to the end users. With up-to-date coverage and extensive treatment of various important topics related to machine learning for fiber-optic communication systems, this book is an invaluable reference for photonics researchers and engineers. It is also a very suitable text for graduate students interested in ML-based signal processing and networking. Discusses the reasons behind the recent popularity of machine learning (ML) concepts in modern optical communication networks and the why/where/how ML can play a unique role Presents fundamental ML techniques like artificial neural networks (ANNs), support vector machines (SVMs), K-means clustering, expectation-maximization (EM) algorithm, principal component analysis (PCA), independent component analysis (ICA), reinforcement learning, and more Covers advanced deep learning (DL) methods such as deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative adversarial networks (GANs) Individual chapters focus on ML applications in key areas of optical communications and networking