Related papers: Energy-Efficient Time-Domain Vector-by-Matrix Mult…
Deep neural networks (DNNs) have revolutionized the field of artificial intelligence and have achieved unprecedented success in cognitive tasks such as image and speech recognition. Training of large DNNs, however, is computationally…
Analog computing based on memristor technology is a promising solution to accelerating the inference phase of deep neural networks (DNNs). A fundamental problem is to map an arbitrary matrix to a memristor crossbar array (MCA) while…
Fast Fourier transform algorithms are an arsenal of effective tools for solving various problems of analysis and high-speed processing of signals of various natures. Almost all of these algorithms are designed to process sequences of…
We demonstrate transfer learning-assisted neural network models for optical matrix multipliers with scarce measurement data. Our approach uses <10\% of experimental data needed for best performance and outperforms analytical models for a…
DNNs are widely used but face significant computational costs due to matrix multiplications, especially from data movement between the memory and processing units. One promising approach is therefore Processing-in-Memory as it greatly…
With power consumption becoming a critical processor design issue, specialized architectures for low power processing are becoming popular. Several studies have shown that neural networks can be used for signal processing and pattern…
In this work, a new energy-efficiency performance metric is proposed for MIMO (multiple input multiple output) point-to-point systems. In contrast with related works on energy-efficiency, this metric translates the effects of using finite…
Silicon offers the enticing opportunity to integrate hybrid quantum-classical computing systems on a single platform. For qubit control and readout, high-frequency signals are required. Therefore, devices that can facilitate its generation…
The contribution of Stream ciphers to cryptography is immense. For fast encryption, stream ciphers are preferred to block ciphers due to their XORing operation, which is easier and faster to implement. In this paper we present a matrix…
The inherent diversity of computation types within the deep neural network (DNN) models often requires a variety of specialized units in hardware processors, which limits computational efficiency, increasing both inference latency and power…
In this paper a low power multiplier is proposed. The proposed multiplier utilizes Broken-Array Multiplier approximation method on the conventional modified Booth multiplier. This method reduces the total power consumption of multiplier up…
Vector modes are spatial modes that have spatially inhomogeneous states of polarization, such as, radial and azimuthal polarization. They can produce smaller spot sizes and stronger longitudinal polarization components upon focusing. As a…
The emerging memristor crossbar array based computing circuits exhibit computing speeds and energy efficiency far surpassing those of traditional digital processors. This type of circuits can complete high-dimensional matrix operations in…
Emerging non-volatile memory (NVM), or memristive, devices promise energy-efficient realization of deep learning, when efficiently integrated with mixed-signal integrated circuits on a CMOS substrate. Even though several algorithmic…
The need for increasingly powerful computing hardware has spawned many ideas stipulating, primarily, the replacement of traditional transistors with alternate "switches" that dissipate miniscule amounts of energy when they switch and…
Processing-in-memory (PIM) has emerged as the go to solution for addressing the von Neumann bottleneck in edge AI accelerators. However, state-of-the-art (SoTA) digital PIM approaches suffer from low compute density, primarily due to the…
This paper proposes and evaluates a novel architecture for a low-power Time-to-Digital Converter with high resolution, optimized for both integration in multichannel chips and high rate operation (40 Mconversion/s/channel). This converter…
In modern neural networks like Transformers, linear layers require significant memory to store activations during backward pass. This study proposes a memory reduction approach to perform backpropagation through linear layers. Since the…
A time-domain analog-weighted-sum calculation model based on a pulse-width modulation (PWM) approach is proposed. The proposed calculation model can be applied to any types of network structure including multi-layer feedforward networks. We…
This paper presents a power distribution network (PDN) decoupling capacitor optimization application with three primary goals: reduction of solution times for large networks, development of flexible network scoring routines, and a…