An optoacoustic field-programmable perceptron for recurrent neural networks

verfasst von: Steven Becker, Dirk Englund, Birgit Stiller
Abstract: Recurrent neural networks (RNNs) can process contextual information such as time series signals and language. But their tracking of internal states is a limiting factor, motivating research on analog implementations in photonics. While photonic unidirectional feedforward neural networks (NNs) have demonstrated big leaps, bi-directional optical RNNs present a challenge: the need for a short-term memory that (i) programmable and coherently computes optical inputs, (ii) minimizes added noise, and (iii) allows scalability. Here, we experimentally demonstrate an optoacoustic recurrent operator (OREO) which meets (i, ii, iii). OREO contextualizes the information of an optical pulse sequence via acoustic waves. The acoustic waves link different optical pulses, capturing their information and using it to manipulate subsequent operations. OREO’s all-optical control on a pulse-by-pulse basis offers simple reconfigurability and is used to implement a recurrent drop-out and pattern recognition of 27 optical pulse patterns. Finally, we introduce OREO as bi-directional perceptron for new classes of optical NNs.
Externe Organisation(en): Max-Planck-Institut für die Physik des Lichts
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU Erlangen-Nürnberg)
Massachusetts Institute of Technology (MIT)
Typ: Artikel
Journal: Nature Communications
Band: 15
Anzahl der Seiten: 8
ISSN: 2041-1723
Publikationsdatum: 16.04.2024
Publikationsstatus: Elektronisch veröffentlicht (E-Pub)
Peer-reviewed: Ja
ASJC Scopus Sachgebiete: Chemie (insg.), Biochemie, Genetik und Molekularbiologie (insg.), Physik und Astronomie (insg.)
Elektronische Version(en): https://doi.org/10.1038/s41467-024-47053-6 (Zugang: Offen)

BibTeX

@article{9fb69482bd074a2c8ed68113a26f2d12,
title = "An optoacoustic field-programmable perceptron for recurrent neural networks",
abstract = "Recurrent neural networks (RNNs) can process contextual information such as time series signals and language. But their tracking of internal states is a limiting factor, motivating research on analog implementations in photonics. While photonic unidirectional feedforward neural networks (NNs) have demonstrated big leaps, bi-directional optical RNNs present a challenge: the need for a short-term memory that (i) programmable and coherently computes optical inputs, (ii) minimizes added noise, and (iii) allows scalability. Here, we experimentally demonstrate an optoacoustic recurrent operator (OREO) which meets (i, ii, iii). OREO contextualizes the information of an optical pulse sequence via acoustic waves. The acoustic waves link different optical pulses, capturing their information and using it to manipulate subsequent operations. OREO{\textquoteright}s all-optical control on a pulse-by-pulse basis offers simple reconfigurability and is used to implement a recurrent drop-out and pattern recognition of 27 optical pulse patterns. Finally, we introduce OREO as bi-directional perceptron for new classes of optical NNs.",
author = "Steven Becker and Dirk Englund and Birgit Stiller",
note = "Publisher Copyright: {\textcopyright} The Author(s) 2024.",
year = "2024",
month = apr,
day = "16",
doi = "10.1038/s41467-024-47053-6",
language = "English",
volume = "15",
journal = "Nature Communications",
issn = "2041-1723",
publisher = "Nature Publishing Group",
number = "1",
}