The Road to Innovation for AI Chips in China
Offered By: tinyML via YouTube
Course Description
Overview
Explore the landscape of AI chip innovation in China through this insightful conference talk from tinyML Asia 2020. Delve into the rapid growth of the Chinese AI chip market and its projected expansion. Examine the emergence of numerous start-up companies competing with international giants in specialized fields. Learn about the academic community's efforts to advance new AI chip technologies. Understand the limitations of traditional computing architectures and the potential of reconfigurable computing technology for flexible, energy-efficient AI solutions. Discover the ongoing research into highly parallel computing architectures with hardware reconfiguration capabilities. Gain insights into various AI chip designs, including dataflow processors, sparse architectures, neuromorphic chips, and reconfigurable systems. Analyze the challenges and opportunities in this evolving field, where innovation in both scientific research and industrial applications continues to thrive.
Syllabus
Introduction
Overview
Classification
AI Chip in China
Complicant Instructions
Dataflow Processor
Kasia Architecture
Resource Utilization
Energy Efficiency
Multisparsity Compatible Commerce
Sparse FCS
Computing Memory
Diphoto Graph
Memory Architecture
Reconfigurable Architecture
Syncr2 Chip
Weak Kernel Decomposition
Overall Architecture
Neuromorphic Chip
Neuron Morphological Devices
Summary
Conclusion
Taught by
tinyML
Related Courses
Stanford Seminar - Neuromorphic Chips - Addressing the Nanostransistor ChallengeStanford University via YouTube All About AI Accelerators - GPU, TPU, Dataflow, Near-Memory, Optical, Neuromorphic & More
Yannic Kilcher via YouTube TinyML Talks - The New Neuromorphic Analog Signal Processor Concept and Technology Platform
tinyML via YouTube Novel Device and Materials in Emerging Memory for Neuromorphic Computing
tinyML via YouTube Make the Signal Chain More Intelligent and Efficient with Mixed Signal Processing and In-Memory Computing
tinyML via YouTube