Hardware Technologies for Artificial Intelligence

AI Chips, Ising Machines, and In-Memory Computing
1. Auflage 2025
ISBN: 978-1-032-98512-1
Verlag: Taylor & Francis Ltd

Buch, Englisch, 230 Seiten, Format (B × H): 156 mm x 234 mm, Gewicht: 453 g

In this comprehensive reference work for researchers, engineers, and students, Kawahara provides a one-stop exploration of next-generation computing at the LSI circuit level, with a focus on the integration of AI, advanced LSI design, Ising machines, and memory innovations.

While current GPUs have high parallel processing capabilities suitable for computations on large datasets, their power consumption is approaching its limit and requires further development. Additionally, edge computing is becoming increasingly important alongside cloud computing. Amid these significant technological trends, this book provides readers with important insights into next-generation computing, namely (1) neural network (artificial intelligence) LSIs and their low power and high performance, (2) hardware design technology for combinatorial optimization problems and Ising machines, and (3) semiconductor memory and data-centric computing. Kawahara first describes the basics of LSI design and neural networks before then describing their large-scale integration, power efficiency and performance enhancements. He then also explains hardware design techniques for Ising Machines, offers case studies of fully coupled Ising Machine LSI. Last, he discusses the basics of semiconductor memory, near/in memory AI computing, and then examines the future prospects. Readers will be able to apply this knowledge to the design and manufacture of such devices to overcome the limitations of current hardware and computational methods, driving future advancements in artificial intelligence and optimization.

This is a valuable reference for students, engineers and researchers alike in this field. As it begins with the basics, it enables all readers to follow the direction of next-generation computing and its important technical content without the need for prior knowledge or reference to other books.

Zielgruppe

Academic, Postgraduate, Professional Reference, and Undergraduate Advanced

Autoren/Hrsg.

Kawahara, Takayuki

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Neuronale Netzwerke
• Technische Wissenschaften Elektronik | Nachrichtentechnik Elektronik
• Mathematik | Informatik EDV | Informatik Technische Informatik
• Mathematik | Informatik EDV | Informatik Computerkommunikation & -vernetzung Cloud-Computing, Grid-Computing

Inhaltsverzeichnis

1. Introduction: AI computing 2. Overview of artificial intelligence hardware LSI (AI chips) and its components 3. Basics of LSI (Large Scale Integrated Circuits) for AI 4. Basic structure of AI chips and various neural networks 5. Low-power, high-performance AI chips structure and related computing 6. Ising machines and combinatorial optimization problem 7. Fully coupled Ising machines (A case study) 8. Semiconductor Memory and Computing 9. Main Semiconductor Memory Features 10. AI computing with semiconductor memory (In-Memory Computing) 11. Future Prospects

Über Autor(innen)

Takayuki Kawahara is a professor at Tokyo University of Science. He earned his Bachelor’s, Master’s and Doctorate from Kyushu University in 1983, 1985, and 1993, respectively. He has significant experience within both industry and academia and is a member of the IEICE and a fellow of the IEEE.