In the context of the METASPIN project we are offering one or more:
PhD Position – Continual Learning with Metaplastic Neural Networks and Nanodevices
The Peter Grünberg Institute - Neuromorphic Software Ecosystems (PGI-15) led by Prof. Dr. Emre Neftci explores neuromorphic computing technologies that learn and work like the brain. PGI-15 research aims to design computing algorithms and architectures from the perspective of neuroscience and physical realizations in dedicated circuits and devices. Applications of our research are targeted to improve computing in many areas, including computer vision, autonomous control, data processing in mobile and wearable devices, and (neuro-)scientific data analysis. A new European project, METASPIN, envisions a radically new low-power artificial synapse technology based on spintronic nanodevices that will prevent catastrophic forgetting, i.e. the loss of memory of previously learned tasks upon learning new ones. This project will develop a new class of device that will use magneto-ionics to support synaptic metaplasticity, i.e. a feature inspired by the human brain based on assigning a ‘hidden value’ encoding its important to the task. This value will affect how ‘plastic’ synaptic state is, giving a hierarchy to previously learned information and thus preventing catastrophic forgetting. The ultimate goal of this project is to integrate this advanced synaptic technology and learning algorithms into an ANN demonstrator to test. This project is in collaboration with Université Paris-Saclay and the University of Zürich.
Last Modified: 31.01.2023