Utkarsh Singhal

TL;DR: An approach for selecting problem-specific Fourier features that greatly improves the performance of MLPs for low-dimensional regression tasks relevant to the computer vision and graphics communities is suggested.

TL;DR: Neural dust is demonstrated, a wireless and scalable ultrasonic backscatter system for powering and communicating with implanted bioelectronics and passive, battery-less communication using backscatter enables high-fidelity transmission of electromyogram (EMG) and electroneurogram (ENG) signals from anesthetized rats.

TL;DR: In this paper, Fourier feature mapping is used to transform the effective neural tangent kernel into a stationary kernel with a tunable bandwidth, which greatly improves the performance of multilayer perceptrons for low-dimensional regression tasks.

TL;DR: The increasing popularity of mobile devices such as smart phones in applications including smart payments and personal health sets a pressing need for improved security without compromised ease of use, and fingerprint recognition has emerged as a particularly attractive option.

TL;DR: A miniaturized portable ultrasonic imager that uses a custom ASIC and a piezoelectric transducer array to transmit and capture 2-D sonographs that can image human tissue as deep as 5 cm while consuming less than 16.5 μJ per pulse-echo measurement.