Analog Neuromorphic Processors

Erscheinungsjahr 2027
ISBN: 978-3-032-33148-9
Verlag: Springer Nature Switzerland AG

Buch, Englisch, 200 Seiten, Format (B × H): 155 mm x 235 mm

This book explores the intersection between analog signals, neuromorphic systems, and computing processors. Here, we find a class of computing devices able to solve biological or deep learning tasks by setting up an analog model of the system of interest. While graphics processing units played a pivotal role in enabling the recent deep learning breakthrough, the advancement of the AI era is posing a pressing demand for novel computing, communications, and information processing capabilities. This shifting landscape has sparked a renaissance in analog processors. Far from trying to replace digital CPUs, this specialized hardware is designed to complement them — pushing the boundaries of computing by (I) enabling new computing paradigms or (II) achieving levels of performance that are otherwise unattainable.

Divided into two parts, the book discusses what analog computing is and why it has the potential to be the key enabling technology for the next generation of AI hardware. The first part discusses the fundamental concepts necessary to understand analog neuromorphic processors, from an introduction to modern AI and its models, to the discussion of its shortcomings and ending with an in-depth discussion on analog signals. The second part presents electronic and photonic analog architectures encompassing memristor arrays, spiking neurons, and reservoir computing. This book aims to introduce AI and analog devices at various levels of analysis to anyone interested in the field, from science enthusiasts to graduate students and professionals. Its structured and pedagogical approach also makes it a prime candidate for textbook adoption in suitable university-level courses.

Zielgruppe

Graduate

Autoren/Hrsg.

De Marinis, Lorenzo

Fachgebiete

• Technische Wissenschaften Elektronik | Nachrichtentechnik Nachrichten- und Kommunikationstechnik Signalverarbeitung
• Technische Wissenschaften Maschinenbau | Werkstoffkunde Technische Mechanik | Werkstoffkunde
• Naturwissenschaften Physik Elektromagnetismus Optik
• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz
• Technische Wissenschaften Elektronik | Nachrichtentechnik Elektronik
• Naturwissenschaften Physik Thermodynamik Festkörperphysik, Kondensierte Materie

Inhaltsverzeichnis

Introduction: Computing, Analog systems, and Neuromorphic Devices.- Fundamental Concepts: Artificial Intelligence and Analog Computing.- Introduction to Modern AI.- Networks of Neurons.- Training Deep Networks.- Modern AI Challenges.- Analog Signals, Noise, and Distortions.- Electronic and Photonic Processors.- Memristor Crossbar Arrays.- Electronic Architectures.- Photonics Fundamentals.- Photonic Architectures.

Über Autor(innen)

Lorenzo De Marinis is an Assistant Professor at the Institute of Telecommunications, Computer Science and Photonics at the Sant'Anna School of Advanced Studies in Pisa, Italy. In 2022, he received his Ph.D. cum laude from Scuola Superiore Sant’Anna, with a doctoral thesis dedicated to Integrated Photonic Neuromorphic Computing. His primary research activities focus on the design, optimization, and validation of innovative photonic integrated circuits for neuromorphic computing, analog processing, and quantum applications. With a background in electronic engineering, he also manages the electro-optic codesign of photonic systems and sub-systems. His broader research interests also encompass network security and distributed computing.