Momchil Minkov
Auto-generated profile, based on publicly available information

Momchil Minkov

Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 December 2019 in Science Advances
Verified
Journal Article
Open access
PDF file
THTyler W. Hughes
Tyler W. Hughes
IWIan A. D. Williamson
Ian A. D. Williamson
MMMomchil Minkov
Momchil Minkov
SFShanhui Fan
Shanhui Fan
Analog machine learning computations are performed passively by propagating light and sound waves through programmed materials.

28 April 2023 in Science
Verified
Journal Article
Closed access
SPSunil Pai
Sunil Pai
ZSZhanghao Sun
Zhanghao Sun
THTyler W. Hughes
Tyler W. Hughes
TPTaewon Park
Taewon Park
BBBen Bartlett
Ben Bartlett
DMDavid A. B. Miller
David A. B. Miller
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 ( > 94% test accuracy), and energy scaling analysis indicated a route to scalable machine learning.

3 December 2018 in ACS Photonics
Verified
Journal Article
Closed access
THTyler W. Hughes
Tyler W. Hughes
MMMomchil Minkov
Momchil Minkov
IWIan A. D. Williamson
Ian A. D. Williamson
SFShanhui Fan
Shanhui Fan

3 January 2020 in Science
Verified
Journal Article
Closed access
PDF file
ADAvik Dutt
Avik Dutt
QLQian Lin
Qian Lin
LYLuqi Yuan
Luqi Yuan
MMMomchil Minkov
Momchil Minkov
MXMeng Xiao
Meng Xiao
SFShanhui Fan
Shanhui Fan
Optically contorting into new dimensions: Creating synthetic dimensions has generated interest in many branches of science, ranging from ultracold atomic physics to photonics. The ability to do so provides a versatile platform for realizing effective gauge potentials and novel topological physics that might be difficult or impossible to realize in real systems. Dutt et al. show that a structured optical ring cavity can sustain more than one synthetic dimension. Under modulation, coupling the different degrees of freedom within the resonator is used synthesize two additional dimensions. The authors are then able to emulate many complex physical phenomena usually associated with condensed matter systems. Science , this issue p. 59

23 July 2019 in IEEE Journal of Selected Topics in Quantum Electronics
Verified
Journal Article
Closed access
IWIan A. D. Williamson
Ian A. D. Williamson
THTyler W. Hughes
Tyler W. Hughes
MMMomchil Minkov
Momchil Minkov
BBBen Bartlett
Ben Bartlett
SPSunil Pai
Sunil Pai
SFShanhui Fan
Shanhui Fan
We introduce an electro-optic hardware platform for nonlinear activation functions in optical neural networks. The optical-to-optical nonlinearity operates by converting a small portion of the input optical signal into an analog electric signal, which is used to intensity -modulate the original optical signal with no reduction in processing speed. Our scheme allows for complete nonlinear on-off contrast in transmission at relatively low optical power thresholds and eliminates the requirement of having additional optical sources between each of the layers of the network Moreover, the activation function is reconfigurable via electrical bias, allowing it to be programmed or trained to synthesize a variety of nonlinear responses. Using numerical simulations, we demonstrate that this activation function significantly improves the expressiveness of optical neural networks, allowing them to perform well on two benchmark machine learning tasks: learning a multi-input exclusive-OR (XOR) logic function and classification of images of handwritten numbers from the MNIST dataset. The addition of the nonlinear activation function improves test accuracy on the MNIST task from 85% to 94%.

9 June 2020 in ACS Photonics
Verified
Journal Article
Closed access
MMMomchil Minkov
Momchil Minkov
IWIan A. D. Williamson
Ian A. D. Williamson
LALucio C. Andreani
Lucio C. Andreani
DGDario Gerace
Dario Gerace
BLBeicheng Lou
Beicheng Lou
SFShanhui Fan
Shanhui Fan