Published October 24, 2024 | Version v1
Dataset Open

Data for: Neuromorphic weighted sums with magnetic skyrmions

Description

Description

The following experimental data were obtained on lithography devices made of magnetic multilayer tracks and thin tantalum transverse electrodes by Kerr microscopy and anomalous Hall effect measurements. The results, demonstrating the weighted sum operation using magnetic skyrmions, are published in T. da Câmara Santa Clara Gomes et al., Neuromorphic weighted sums with magnetic skyrmions, Nature Electronics (2024). Please find in the README additional information regarding the data files and the variables.

Abstract

Integrating magnetic skyrmions into neuromorphic computing could help improve hardware efficiency and computational power. However, developing a scalable implementation of the weighted sum of neuron signals — a core operation in neural networks — has remained a challenge. Here, we show that weighted sum operations can be performed in a compact, biologically-inspired manner by using the non-volatile and particle-like characteristics of magnetic skyrmions that make them easily countable and summable. The skyrmions are electrically generated in numbers proportional to the input with an efficiency given by a non-volatile weight. The chiral particles are then directed using localized current injections to a location where their presence is quantified through non-perturbative electrical measurements. Our experimental demonstration, which currently has two inputs, can be scaled to accommodate multiple inputs and outputs using a crossbar array design, potentially nearing the energy efficiency observed in biological systems.

Files

Fig2e_Nsk_vs_Npulses_148GA_m2.txt

Files (40.6 kB)

Name Size Download all
md5:61bc177af936e939d6a04f5663032dea
382 Bytes Preview Download
md5:9acb7ea7c7cd8701266550e897d6e2dc
385 Bytes Preview Download
md5:c985f4a8ea62c6681d138f6c7cea5c70
392 Bytes Preview Download
md5:a489cdde9744ffe4d76284e5fd1b5cbf
267 Bytes Preview Download
md5:39348c0e40d68384fdad170a3b8196ef
384 Bytes Preview Download
md5:9b2266cd62229d6ebaa4a38a40a1421e
389 Bytes Preview Download
md5:c985f4a8ea62c6681d138f6c7cea5c70
392 Bytes Preview Download
md5:3ebb6b7378abab7a97b56bd5d0484b6a
395 Bytes Preview Download
md5:fe2784e1e235a955765e1541d83aa903
362 Bytes Preview Download
md5:14faf69aeaede9a2633d898d8e7b86e4
399 Bytes Preview Download
md5:c985f4a8ea62c6681d138f6c7cea5c70
392 Bytes Preview Download
md5:a3236aa23f018d8074735711154780d9
385 Bytes Preview Download
md5:a4beda1a151da54f5fabf028c1d8c42a
165 Bytes Preview Download
md5:573b03d2cd36178f0877b31a39da7557
64 Bytes Preview Download
md5:516f4b1e43929fe378f21b7044271d02
1.1 kB Preview Download
md5:cecbed9f055902c66d7978d01c393de1
1.1 kB Preview Download
md5:a4bfd00dcaabc0bd0d51fbc035693bc9
1.1 kB Preview Download
md5:5b98a1e772c3cec9d3d26674b3846d31
3.2 kB Preview Download
md5:96fc5c5372a5d1ba5a86abe9cd124e5a
551 Bytes Preview Download
md5:327b4a4f1371e926f88a7861d0a8c304
25 Bytes Preview Download
md5:7d65e54767ba5c40cc8f8c6d25e63790
3.8 kB Preview Download
md5:5c434c382480c878a2a374549bebe699
162 Bytes Preview Download
md5:2913b610837aaa45b6271c69c55945cb
3.8 kB Preview Download
md5:46b7c35ef15ca4e838a473ae744457d0
164 Bytes Preview Download
md5:84a6e0d162bb6463d07a29b9b667d4af
210 Bytes Preview Download
md5:21f50715fe8e4e0980a9d2e0729cedbe
210 Bytes Preview Download
md5:5619ba5bce2b2dd56c9f64eacc84564e
210 Bytes Preview Download
md5:cc6227ab4e6fef2e39970ea59556937f
210 Bytes Preview Download
md5:33a2fc147629298085ec71bc5e6c825f
7.7 kB Preview Download
md5:38dcf1d77a5b7b2d1edea945b93830bf
7.7 kB Preview Download
md5:13c8d67093cacff559d4cededede0655
4.5 kB Preview Download

Additional details

Related works

Is published in
Preprint: arXiv:2310.16909 (arXiv)
Publication: 10.1038/s41928-024-01303-z (DOI)

Funding

European Commission
FET-Proactive Grant SKYTOP 824123
European Research Council
GrenaDyn 101020684
European Union
SkyANN 101135729
Agenția Națională pentru Cercetare și Dezvoltare
PEPR SPIN CHIREX ANR-22-EXSP-0002