EUVIMEDEUVIMED entdecken
Crossref

Magnetic-Tunnel-Junction-Based Probabilistic Processor for Efficient Embodied Evolution

Youwei Bao, Shuhan Yang, Edward Humianto, Hyunsoo Yang iD

Nano Letters · 06.05.2026 · 26 (17) · 5731-5739

DOI: 10.1021/acs.nanolett.6c00275 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Nano Letters, online erschienen am 22.04.2026. Verfasst von Youwei Bao und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electrical and Computer Engineering, National University of Singapore, Singapore 117583, Singapore. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 61 Literaturangaben. Gefördert von: Ministry of Education - Singapore, National Research Foundation Singapore.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In embodied artificial intelligence (AI), evolutionary search enables adaptation in complex and uncertain environments but relies on massive stochastic sampling, which in hardware is typically generated using complementary metal-oxide-semiconductor pseudorandom number generators with significant power and area costs. In this paper, we present a 256-magnetic-tunnel-junction (MTJ)-based probabilistic processor as an efficient physical platform for embodied evolution. Using an Ising-based probabilistic sampler, we experimentally construct highly reconfigurable Gaussian probabilistic bits from networks of MTJ-based probabilistic bits and scale the architecture to 256 parallel generators for evolutionary mutation. We demonstrate evolved agent morphologies comparable to those obtained using an ideal software Gaussian random number generator in a robotic locomotion task. MTJ-based stochasticity further drives on-hardware selection, including a diversity-aware scheme that balances lineage diversity and fitness. Our results establish that MTJ-based probabilistic hardware is a promising low-power stochastic substrate for embodied AI and evolutionary robotics.

Schlagwörter

Stichwörter der Autor:innen

embodied artificial intelligencemagnetic tunnel junctionsprobabilistic computing hardwarereconfigurable random number generator

Autor:innen und Institutionen

  • Department of Electrical and Computer Engineering, National University of Singapore, Singapore 117583, Singapore.

ORCID-Profile: Hyunsoo Yang

Förderung

  • Ministry of Education - Singapore
  • National Research Foundation Singapore

Zitieren

APA

Bao, Y., Yang, S., Humianto, E., & Yang, H. (2026). Magnetic-Tunnel-Junction-Based Probabilistic Processor for Efficient Embodied Evolution. Nano Letters, 26(17), 5731-5739. https://doi.org/10.1021/acs.nanolett.6c00275

Vancouver

Bao Y, Yang S, Humianto E, Yang H. Magnetic-Tunnel-Junction-Based Probabilistic Processor for Efficient Embodied Evolution. Nano Letters. 2026;26(17):5731-5739. doi:10.1021/acs.nanolett.6c00275

BibTeX
@article{Bao2026,
  author = {Bao, Youwei and Yang, Shuhan and Humianto, Edward and Yang, Hyunsoo},
  title = {Magnetic-Tunnel-Junction-Based Probabilistic Processor for Efficient Embodied Evolution},
  journal = {Nano Letters},
  year = {2026},
  volume = {26},
  number = {17},
  pages = {5731--5739},
  doi = {10.1021/acs.nanolett.6c00275}
}
Kostenlos starten →