Mixed-mode in-memory computing using adaptive phase-change materials (C06)

• Subject Area: Experimental Condensed Matter Physics ; Optics, Quantum Optics and Physics of Atoms, Molecules and Plasmas
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 433682494

In this project, we employ mixed-mode neuromorphic architectures to implement in-memory learning through electrically programmable artificial neurons and synapses based on phase-change materials. Optimal compositions will be devised, implemented, and tested using advanced fabrication methods and nanoscale analytics. We will develop and implement mixed-precision algorithms for in-memory learning and will apply our architecture for processing of external sensory input. In the long term, we strive to develop brain-inspired nanoscale computing devices, which are able to carry out adaptive optical information processing.

• DFG Programme: Collaborative Research Centres
• Subproject of: SFB 1459: Intelligent Matter – From responsive to adaptive nanosystems
• Applicant Institution: Universität Münster
• Project Heads: Professor Dr. Wolfram Hans Peter Pernice; Professor Dr. Martin Salinga