Tantárgy adatlapja
The objective of the course is to introduce analog neuromorphic computing to the students, with the required device physics, circuit design background and to give an overview of analog models of computation. Most important topics include:
Energy and power consumption of digital logic
Implementation of neural networks by digital circuits
Basics of transistor operation
Simple digital circuits from transistors
Operational amplifiers, transconductance amplifiers, oscillators
Emerging devices, memristors, spin-based devices
Integrator and differentiator circuits, analog filters
Switched capacitor circuits
Design of a CNN cell, CNN circuits
Design of artificial analog retinas
Design of coupled oscillator circuits
Spiking neural network design
Hardware accelerators for convolutional neural networks
Sensors and amplifiers for sensory circuits
List of selected literature:
Mead, Carver. "Neuromorphic electronic systems." Proceedings of the IEEE 78, no. 10 (1990): 1629-1636. Kramer, Jörg; Delbrück, Tobias; Liu, Shih Chi; Indiveri, Giacomo: Computation in neuromorphic analog VLSI systems https://doi.org/10.3929/ethz-a-004306354 Horowitz, Paul, and Winfield Hill. The art of electronics. Cambridge: Cambridge university press, 1978. Rahul Sarpeshkar: Ultra Low Power Bioelectronics: Fundamentals, Biomedical Applications, and Bio-Inspired Systems Cambridge University Press; 1st edition (February 22, 2010) List of those required professional competences, competence elements to the development of which the subject characteristically, materially contributes: a) knowledge of new algorithms, circuit design methods, device physics b) skills individual work, design
