Dresden 2020 – wissenschaftliches Programm
Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
O: Fachverband Oberflächenphysik
O 68: PhD Focus Session: Symposium on "Magnetism – A Potential Platform for Big Data?" (joint session MA/O/AKjDPG)
O 68.4: Hauptvortrag
Mittwoch, 18. März 2020, 16:45–17:15, HSZ 04
Unconventional computing with stochastic magnetic tunnel junctions — •Alice Mizrahi1, 2, 3,4, Tifenn Hirtzlin2, Matthew Daniels3, 4, Nicolas Locatelli2, Akio Fukushima5, Hit Kubota5, Shinsi Yuasa5, MD Stiles4, Julie Grollier1, and Damien Querlioz2 — 1Unité Mixte de Physique CNRS, Thales, Univ. Paris-Sud, Université Paris-Saclay, 91767 Palaiseau, France — 2Centre de Nanosciences et de Nanotechnologies, Univ. Paris-Sud, CNRS, Université Paris-Saclay, 91405, Orsay, France — 3Maryland NanoCenter, University of Maryland, College Park, Maryland 20742, USA — 4National Institute of Standards and Technology, Gaithersburg, Maryland 20899, USA — 5Spintronics Research Center, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, 305-8568, Japan
Magnetic tunnel junctions are bi-stable nanodevices which magnetic state can be both read and written electrically. Their high endurance, reliability and CMOS-compatibility have made them flagship devices for novel forms of computing. While they are mostly used as non-volatile binary memories, they can be made unstable and thus behave as stochastic oscillators. Here, we show how stochastic magnetic tunnel junctions are promising elements for low energy implementations of unconventional computing. An analogy can be drawn between stochastic magnetic tunnel junctions and stochastic spiking neurons. We apply neuroscience computing paradigm to these devices and demonstrate that they can be the building blocks of low energy artificial neural networks capable of on-chip learning.