Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367173722> ?p ?o ?g. }
- W4367173722 endingPage "404" @default.
- W4367173722 startingPage "398" @default.
- W4367173722 abstract "Integrated photonic neural networks provide a promising platform for energy-efficient, high-throughput machine learning with extensive scientific and commercial applications. Photonic neural networks efficiently transform optically encoded inputs using Mach-Zehnder interferometer mesh networks interleaved with nonlinearities. We experimentally trained a three-layer, four-port silicon photonic neural network with programmable phase shifters and optical power monitoring to solve classification tasks using “in situ backpropagation,” a photonic analog of the most popular method to train conventional neural networks. We measured backpropagated gradients for phase-shifter voltages by interfering forward- and backward-propagating light and simulated in situ backpropagation for 64-port photonic neural networks trained on MNIST image recognition given errors. All experiments performed comparably to digital simulations ( <mml:math xmlns:mml=http://www.w3.org/1998/Math/MathML overflow=scroll> <mml:mo>></mml:mo> </mml:math> 94% test accuracy), and energy scaling analysis indicated a route to scalable machine learning." @default.
- W4367173722 created "2023-04-28" @default.
- W4367173722 creator A5001912998 @default.
- W4367173722 creator A5006242251 @default.
- W4367173722 creator A5007432815 @default.
- W4367173722 creator A5020714759 @default.
- W4367173722 creator A5043102840 @default.
- W4367173722 creator A5044137053 @default.
- W4367173722 creator A5051199045 @default.
- W4367173722 creator A5058424108 @default.
- W4367173722 creator A5058799196 @default.
- W4367173722 creator A5068174848 @default.
- W4367173722 creator A5070817954 @default.
- W4367173722 creator A5082030946 @default.
- W4367173722 creator A5090513654 @default.
- W4367173722 creator A5091113379 @default.
- W4367173722 date "2023-04-28" @default.
- W4367173722 modified "2023-10-16" @default.
- W4367173722 title "Experimentally realized in situ backpropagation for deep learning in photonic neural networks" @default.
- W4367173722 cites W1486585911 @default.
- W4367173722 cites W1498436455 @default.
- W4367173722 cites W1575843463 @default.
- W4367173722 cites W1776515882 @default.
- W4367173722 cites W1958979682 @default.
- W4367173722 cites W1999085092 @default.
- W4367173722 cites W2007339694 @default.
- W4367173722 cites W2018435387 @default.
- W4367173722 cites W2067427618 @default.
- W4367173722 cites W2076476236 @default.
- W4367173722 cites W2078357199 @default.
- W4367173722 cites W2097100165 @default.
- W4367173722 cites W2108199536 @default.
- W4367173722 cites W2108598243 @default.
- W4367173722 cites W2151672110 @default.
- W4367173722 cites W2168648327 @default.
- W4367173722 cites W2203188256 @default.
- W4367173722 cites W2537188330 @default.
- W4367173722 cites W2740400597 @default.
- W4367173722 cites W2752849906 @default.
- W4367173722 cites W2754492346 @default.
- W4367173722 cites W2904694042 @default.
- W4367173722 cites W2905425616 @default.
- W4367173722 cites W2952285760 @default.
- W4367173722 cites W2962730419 @default.
- W4367173722 cites W2964082513 @default.
- W4367173722 cites W2974263490 @default.
- W4367173722 cites W2976859966 @default.
- W4367173722 cites W2982199251 @default.
- W4367173722 cites W3018394831 @default.
- W4367173722 cites W3021423154 @default.
- W4367173722 cites W3030590996 @default.
- W4367173722 cites W3091851330 @default.
- W4367173722 cites W3094052952 @default.
- W4367173722 cites W3098737103 @default.
- W4367173722 cites W3101439286 @default.
- W4367173722 cites W3102122242 @default.
- W4367173722 cites W3103046660 @default.
- W4367173722 cites W3103448394 @default.
- W4367173722 cites W3104406190 @default.
- W4367173722 cites W3105148358 @default.
- W4367173722 cites W3106404940 @default.
- W4367173722 cites W3120016950 @default.
- W4367173722 cites W3121141908 @default.
- W4367173722 cites W3135320274 @default.
- W4367173722 cites W3155628184 @default.
- W4367173722 cites W3163879263 @default.
- W4367173722 cites W3169435745 @default.
- W4367173722 cites W3205615004 @default.
- W4367173722 cites W4210678178 @default.
- W4367173722 cites W4226293750 @default.
- W4367173722 cites W4283819607 @default.
- W4367173722 cites W4308069143 @default.
- W4367173722 cites W4308514394 @default.
- W4367173722 cites W4312968147 @default.
- W4367173722 cites W4313639775 @default.
- W4367173722 cites W4327943790 @default.
- W4367173722 doi "https://doi.org/10.1126/science.ade8450" @default.
- W4367173722 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37104594" @default.
- W4367173722 hasPublicationYear "2023" @default.
- W4367173722 type Work @default.
- W4367173722 citedByCount "13" @default.
- W4367173722 countsByYear W43671737222023 @default.
- W4367173722 crossrefType "journal-article" @default.
- W4367173722 hasAuthorship W4367173722A5001912998 @default.
- W4367173722 hasAuthorship W4367173722A5006242251 @default.
- W4367173722 hasAuthorship W4367173722A5007432815 @default.
- W4367173722 hasAuthorship W4367173722A5020714759 @default.
- W4367173722 hasAuthorship W4367173722A5043102840 @default.
- W4367173722 hasAuthorship W4367173722A5044137053 @default.
- W4367173722 hasAuthorship W4367173722A5051199045 @default.
- W4367173722 hasAuthorship W4367173722A5058424108 @default.
- W4367173722 hasAuthorship W4367173722A5058799196 @default.
- W4367173722 hasAuthorship W4367173722A5068174848 @default.
- W4367173722 hasAuthorship W4367173722A5070817954 @default.
- W4367173722 hasAuthorship W4367173722A5082030946 @default.
- W4367173722 hasAuthorship W4367173722A5090513654 @default.
- W4367173722 hasAuthorship W4367173722A5091113379 @default.
- W4367173722 hasBestOaLocation W43671737222 @default.