VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.

227-0666-00L 3 Credits MSC D-MATL , D-ITET
You're viewing possible stale or outdated data. Please check the latest semester for more up-to-date information.

Neuromorphic Electronics with Oxides: from Materials to AI Hardware

Lecturers & Examiners: Prof. Dr. Laura Bégon-Lours
VVZ CR n/a

Last Updated: 2026-06-01 11:33:22

Abstract

At the interface between materials sciences, electrical engineering and neurosciences, this course presents how the physical properties of materials can be used in electronic circuits which behave like the brain. For example, the quantum tunneling current through a ferroelectric junction can be reversibly increased by flipping the polarization, mimicking the “potentiation” of a biological synapse.

Objective

The students will learn about the properties of novel electronic devices that are required to build AI hardware. The objective is that at the end of the course, the student can propose an application after assessing the electrical characterization results of an electronic device. The student will be able to explain the mechanisms governing the functionality of artificial synapses and neurons, at the materials level.

Content

Artificial Intelligence (AI) workloads such as Natural Language Processing tools or Image Classification process large volumes of data. On conventional “Von Neumann” computers, this data must be transferred between the memory and the processor, dissipating a large amount of energy. Solid-state physics offer a wide range of applications: among them, the research on neuromorphic devices and circuits aims at the development of “Beyond Von Neumann” hardware to support AI. Applications of Neuromorphic technologies In Memory Computing Convolutional Neural Networks + Inference Deep Neural Networks "Online-Learning" in Artificial Neural Network accelerators Invited Lecture: Bert Jan Offrein (IBM Research Zurich) Short and Long Term Memory Bio-inspired synapses with Oxides Nanofabrication techniques Binnig and Rohrer Nanotechnology Center visit (limited number of places) Reinforcement learning Neurons Neuromorphic Sensing with Oxides State-of-the-Art: demonstrators

General Information

Language
English
Levels
MSC
Frequency
Yearly recurring

Examination

Type
end-of-semester examination
Mode
oral 40 minutes
20 min preparation + 20 min oral examination

Course Components

Type Title Time & Place Hours
lecture Neuromorphic Electronics with Oxides: from Materials to AI Hardware
  • Wed 14:15-16:00 (HG D 3.3)
  • 19.02 Date 14:15-16:00 (HG D 1.2)
  • 26.03 Date 14:15-16:00 (HG D 1.2)
2 h weekly
independent project Neuromorphic Electronics with Oxides: from Materials to AI Hardware No time listed 1 h weekly

Offered In