仿神經形態積體電路與系統

Neuromorphic Integrated Circuits and Systems

學期

114-2

學分

3 學分

當期課號

535111

永久課號

EEEE30048

開課單位

電機工程學系

授課教師

彭盛裕

校區

光復

類別

選修

上課時間表

節	週一
N 12:20–13:10	仿神經形態積體電路與系統 EE220 3 節連堂
5 13:20–14:10
6 14:20–15:10

* 根據陽明交大上課時間表所列

概述

Upon completion of this course, students will be able to: Bridge Neuroscience and Silicon Physics Understand the physical isomorphism between biological ion channels and subthreshold MOSFETs to design "bio-plausible" circuits. Master Low-Power Analog Design Acquire proficiency in designing subthreshold current-mode circuits to implement silicon neurons and neuromorphic circuits with ultra-low power consumption. Connect Algorithms to Hardware Analyze the evolution from traditional deep learning architectures and computing-in-memory paradigms to event-driven spiking neural networks. Design Bio-Inspired Sensory Systems Develop integrated neuromorphic systems, such as Silicon Retinas, Silicon Cochlea, and tactile interfaces, to address perception and control challenges in robotics.

先修科目

Microelectronics, Analog Circuit Design, Artificial Intelligence

教學方式

Lectures with Projects and Paper Presentations

評分方式

Homework Assignments 30% Projects 25% Final Exam 25% Paper Presentation 10% Class Participation 10%

週次計畫

週次	主題
第 1 週	Introduction to AI History and Challenges
第 2 週	AI Landscape and Neuromorphic Engineering Basics
第 3 週	Recap on MOS FET Bulk-Referenced Models
第 4 週	Subthreshold MOS & Current-mode Circuits
第 5 週	Translinear & Log-domain Filters
第 6 週	AI Accelerators: Digital vs. Mixed-Signal
第 7 週	National Holiday
第 8 週	AI Accelerators: Digital vs. Mixed-Signal
第 9 週	Emerging Memories & Computing-In-Memories
第 10 週	Silicon Neuron Circuits & AER & Large-scale Architectures
第 11 週	Synaptic Circuits & Plasticity & SNN Algorithms
第 12 週	Visual Pathway & Silicon Retina
第 13 週	Auditory Pathway & Silicon Cochlea /Tactile, Olfactory & Sensor Fusion
第 14 週	Invited Talks/Paper Discussion
第 15 週	Final Project Presentation
第 16 週	Final Exam

教科書

Analog VLSI: Circuits and Principles Authors: Liu, Kramer, Indiveri, Delbruck, and Douglas Publisher: Massachusetts Institute of Technology Year: 2002 Event-Based Neuromorphic Systems Authors: Shih-Chii Liu, Tobi Delbruck, Giacomo Indiveri, Adrian Whatley, Rodney Douglas Publisher: John Wiley & Sons Year: 2015

Office Hours

地點: EE-222B
時間: Monday 11:10am to 12:00pm
聯絡方式: speng@nycu.edu.tw