Error analysis of a 3-Layer SOA-based photonic deep neural network for image classification

Bin Shi; Nicola Calabretta; Ripalta  Stabile

Error analysis of a 3-Layer SOA-based photonic deep neural network for image classification

Bin Shi, Nicola Calabretta, Ripalta Stabile

Research output: Contribution to conference › Paper › Academic

Abstract

We demonstrate Iris flower classification with a trained 3-layer photonic deep neural network, which is implemented by a monolithically integrated SOA-based InP cross-connect chip, achieving a final accuracy of 85.8%. An in-depth analysis of the error evolution in the system suggests that the E/O conversions from layer to layer contribute to the majority of the error.

Original language	English
Publication status	Published - 2019
Event	24th Annual Symposium of the IEEE Photonics Society Benelux - VU Main Building, Amsterdam, Netherlands Duration: 21 Nov 2019 → 22 Nov 2019 Conference number: 24 https://www.aanmelder.nl/ieee-benelux-2019 http://photonics-benelux.org/

Conference

Conference	24th Annual Symposium of the IEEE Photonics Society Benelux
Country/Territory	Netherlands
City	Amsterdam
Period	21/11/19 → 22/11/19
Internet address	https://www.aanmelder.nl/ieee-benelux-2019 http://photonics-benelux.org/

Cite this

@conference{cbfe7b893eb243a08784562a4707c969,

title = "Error analysis of a 3-Layer SOA-based photonic deep neural network for image classification",

abstract = "We demonstrate Iris flower classification with a trained 3-layer photonic deep neural network, which is implemented by a monolithically integrated SOA-based InP cross-connect chip, achieving a final accuracy of 85.8%. An in-depth analysis of the error evolution in the system suggests that the E/O conversions from layer to layer contribute to the majority of the error.",

author = "Bin Shi and Nicola Calabretta and Ripalta Stabile",

year = "2019",

language = "English",

note = "24th Annual Symposium of the IEEE Photonics Society Benelux ; Conference date: 21-11-2019 Through 22-11-2019",

url = "https://www.aanmelder.nl/ieee-benelux-2019, http://photonics-benelux.org/",

}

TY - CONF

T1 - Error analysis of a 3-Layer SOA-based photonic deep neural network for image classification

AU - Shi, Bin

AU - Calabretta, Nicola

AU - Stabile, Ripalta

N1 - Conference code: 24

PY - 2019

Y1 - 2019

N2 - We demonstrate Iris flower classification with a trained 3-layer photonic deep neural network, which is implemented by a monolithically integrated SOA-based InP cross-connect chip, achieving a final accuracy of 85.8%. An in-depth analysis of the error evolution in the system suggests that the E/O conversions from layer to layer contribute to the majority of the error.

AB - We demonstrate Iris flower classification with a trained 3-layer photonic deep neural network, which is implemented by a monolithically integrated SOA-based InP cross-connect chip, achieving a final accuracy of 85.8%. An in-depth analysis of the error evolution in the system suggests that the E/O conversions from layer to layer contribute to the majority of the error.

M3 - Paper

T2 - 24th Annual Symposium of the IEEE Photonics Society Benelux

Y2 - 21 November 2019 through 22 November 2019

ER -

Error analysis of a 3-Layer SOA-based photonic deep neural network for image classification

Abstract

Conference

Fingerprint

Cite this