Search

Your search keyword '"Singh, Alok"' showing total 3,433 results
Start Over Author "Singh, Alok"

Refine your results

more »

more »

more »

more »

more »
3,433 results on '"Singh, Alok"'

15. DNN-GDITD: Out-of-distribution detection via Deep Neural Network based Gaussian Descriptor for Imbalanced Tabular Data

Author

Chudasama, Priyanka, Surisetty, Anil, Malhotra, Aakarsh, and Singh, Alok

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence
Abstract

Classification tasks present challenges due to class imbalances and evolving data distributions. Addressing these issues requires a robust method to handle imbalances while effectively detecting out-of-distribution (OOD) samples not encountered during training. This study introduces a novel OOD detection algorithm designed for tabular datasets, titled Deep Neural Network-based Gaussian Descriptor for Imbalanced Tabular Data (DNN-GDITD). The DNN-GDITD algorithm can be placed on top of any DNN to facilitate better classification of imbalanced data and OOD detection using spherical decision boundaries. Using a combination of Push, Score-based, and focal losses, DNN-GDITD assigns confidence scores to test data points, categorizing them as known classes or as an OOD sample. Extensive experimentation on tabular datasets demonstrates the effectiveness of DNN-GDITD compared to three OOD algorithms. Evaluation encompasses imbalanced and balanced scenarios on diverse tabular datasets, including a synthetic financial dispute dataset and publicly available tabular datasets like Gas Sensor, Drive Diagnosis, and MNIST, showcasing DNN-GDITD's versatility., Comment: 17 pages

Published
2024

21. Giant Thermoelectric Response of Fluxons in Superconductors

Author

Singh, Alok Nath, Bhandari, Bibek, Braggio, Alessandro, Giazotto, Francesco, and Jordan, Andrew N.

Subjects
Condensed Matter - Superconductivity, Condensed Matter - Mesoscale and Nanoscale Physics
Abstract

Thermoelectric devices that operate on quantum principles have been under extensive investigation in the past decades. These devices are at the fundamental limits of miniaturized heat engines and refrigerators, advancing the field of quantum thermodynamics. Most research in this area concerns the use of conduction electrons and holes as charge and heat carriers, and only very recently have superconductors been considered as thermal engines and thermoelectric devices. Here, we investigate the thermoelectric response of an Abrikosov vortex in type-II superconductors in the deep quantum limit. We consider two thermoelectric geometries, a type-II SIN junction and a local Scanning Tunneling Microscope (STM)-tip normal metal probe over the superconductor. We exploit the strong breaking of particle-hole symmetry in bound states at sub-gap energies within the superconducting vortex to realize a giant thermoelectric response in the presence of fluxons. We predict a thermovoltage of a few mV/K at sub-Kelvin temperatures using both semi-analytic and numerical self-consistent solutions of the Bogoliubov-de Gennes equations. Relevant thermoelectric coefficients and figures of merit are found within our model, both in linear and nonlinear regimes. The ZT of the SIN junction is around 1, rising to above 3 for the STM junction centered at the vortex core. We also discuss how this system can be used as a sensitive thermocouple, diode, or localized bolometer to detect low-energy single photons., Comment: 11 pages, 4 figures

Published
2024

26. Inverse Design of Photonic Surfaces via High throughput Femtosecond Laser Processing and Tandem Neural Networks

Author

Park, Minok, Grbčić, Luka, Motameni, Parham, Song, Spencer, Singh, Alok, Malagrino, Dante, Elzouka, Mahmoud, Vahabi, Puya H, Todeschini, Alberto, de Jong, Wibe Albert, Prasher, Ravi, Zorba, Vassilia, and Lubner, Sean D

Subjects
Information and Computing Sciences, Machine Learning, Affordable and Clean Energy, deep learning, femtosecond laser processing, inverse design, machine learning, photonic surface, tandem neural network
Abstract

This work demonstrates a method to design photonic surfaces by combining femtosecond laser processing with the inverse design capabilities of tandem neural networks that directly link laser fabrication parameters to their resulting textured substrate optical properties. High throughput fabrication and characterization platforms are developed that generate a dataset comprising 35280 unique microtextured surfaces on stainless steel with corresponding measured spectral emissivities. The trained model utilizes the nonlinear one-to-many mapping between spectral emissivity and laser parameters. Consequently, it generates predominantly novel designs, which reproduce the full range of spectral emissivities (average root-mean-squared-error < 2.5%) using only a compact region of laser parameter space 25 times smaller than what is represented in the training data. Finally, the inverse design model is experimentally validated on a thermophotovoltaic emitter design application. By synergizing laser-matter interactions with neural network capabilities, the approach offers insights into accelerating the discovery of photonic surfaces, advancing energy harvesting technologies.

Published
2024

Catalog

Books, media, physical & digital resources
See catalog results