تقرير
Machine learning using magnetic stochastic synapses
| العنوان: | Machine learning using magnetic stochastic synapses |
|---|---|
| المؤلفون: | Ellis, Matthew O. A., Welbourne, Alex, Kyle, Stephan J., Fry, Paul W., Allwood, Dan A., Hayward, Thomas J., Vasilaki, Eleni |
| سنة النشر: | 2023 |
| المجموعة: | Computer Science Condensed Matter |
| مصطلحات موضوعية: | Computer Science - Emerging Technologies, Condensed Matter - Mesoscale and Nanoscale Physics, Computer Science - Machine Learning, Computer Science - Neural and Evolutionary Computing |
| الوصف: | The impressive performance of artificial neural networks has come at the cost of high energy usage and CO$_2$ emissions. Unconventional computing architectures, with magnetic systems as a candidate, have potential as alternative energy-efficient hardware, but, still face challenges, such as stochastic behaviour, in implementation. Here, we present a methodology for exploiting the traditionally detrimental stochastic effects in magnetic domain-wall motion in nanowires. We demonstrate functional binary stochastic synapses alongside a gradient learning rule that allows their training with applicability to a range of stochastic systems. The rule, utilising the mean and variance of the neuronal output distribution, finds a trade-off between synaptic stochasticity and energy efficiency depending on the number of measurements of each synapse. For single measurements, the rule results in binary synapses with minimal stochasticity, sacrificing potential performance for robustness. For multiple measurements, synaptic distributions are broad, approximating better-performing continuous synapses. This observation allows us to choose design principles depending on the desired performance and the device's operational speed and energy cost. We verify performance on physical hardware, showing it is comparable to a standard neural network. Comment: 15 pages, 4 figures |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2303.01886 |
| رقم الأكسشن: | edsarx.2303.01886 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |