Publications

Displaying 1 - 8 of 8

Gradient Descent in Recurrent Neural Networks with Model-Free Multiplexed Gradient Descent: Toward Temporal On-Chip Neuromorphic Learning

June 12, 2025

Author(s)

Ryan O'Loughlin, Nicholas Skuda, Bakhrom Oripov, Sonia Buckley, Adam McCaughan

The brain implements recurrent neural networks (RNNs) efficiently, and modern computing hardware does not. Although specialized neuromorphic hardware is well suited for recurrent implementations in the inference phase, it does not lend itself to the

Scaling of hardware-compatible perturbative training algorithms

April 17, 2025

Author(s)

Bakhrom Oripov, Andrew Dienstfrey, Adam McCaughan, Sonia Buckley

In this work, we explore the capabilities of Multiplexed Gradient Descent (MGD), a scalable and efficient perturbative zeroth-order training method for estimating the gradient of a loss function in hardware and training it via stochastic gradient descent

Photonic Online Learning

January 9, 2024

Author(s)

Sonia Buckley, Adam McCaughan, Bakhrom Oripov

Training in machine learning necessarily involves more operations than inference only, with higher precision, more memory, and added computational complexity. In hardware, many implementations side-step this issue by designing "inference-only" hardware

Integration of vertically-aligned carbon nanotubes with superconducting nanowire single photon detectors

November 27, 2023

Author(s)

Dana Rampini, Chris Yung, Bakhrom Oripov, Solomon I. Woods, John H. Lehman, Sae Woo Nam, Adam McCaughan

We report on initial fabrication efforts in the integration of superconducting nanowire single-photon detectors (SNSPDs) with vertically aligned carbon nanotubes (VACNTs) with the goal of creating a wideband single-photon detector. SNSPDs provide high

A superconducting nanowire single-photon camera with 400,000 pixels

October 25, 2023

Author(s)

Bakhrom Oripov, Dana Rampini, Jason Allmaras, Matt Shaw, Sae Woo Nam, Boris Korzh, Adam McCaughan

For the past 50 years, superconducting detectors have offered exceptional sensitivity and speed for detecting faint electromagnetic signals in a wide range of applications. These detectors operate at very low temperatures and generate a minimum of excess

Cryogenic reconfigurable logic with superconducting heater cryotron: Enhancing area efficiency and enabling camouflaged processors

October 10, 2023

Author(s)

Shamiul Alam, Dana Rampini, Bakhrom Oripov, Adam McCaughan, Ahmedullah Aziz

Superconducting electronics are among the most promising alternatives to conventional CMOS technology, thanks to the ultra-fast speed and ultra-high energy efficiency of the superconducting devices. Having a cryogenic control processor is also a crucial

Multiplexed gradient descent: Fast online training of modern datasets on hardware neural networks without backpropagation

June 26, 2023

Author(s)

Adam McCaughan, Bakhrom Oripov, Natesh Ganesh, Sae Woo Nam, Andrew Dienstfrey, Sonia Buckley

We show that model-free perturbative methods can be used to efficiently train modern neural network architectures in a way that can be directly applied to emerging neuromorphic hardware. These methods were investigated for training VLSI neural networks

The thermally-coupled imager: A scalable readout architecture for superconducting nanowire single photon

June 7, 2023

Author(s)

Adam McCaughan, Yao Zhai, Boris Korzh, Jason Allmaras, Bakhrom Oripov, Matthew Shaw, Sae Woo Nam

Although superconducting nanowire single-photon detectors (SNSPDs) are a promising technology for quantum optics, metrology, and astronomy, they currently lack a readout architecture that is scalable to the megapixel regime and beyond. In this work, we

Search Publications by: Bakhrom Oripov (IntlAssoc)