Related papers: Temperature-Insensitive Analog Vector-by-Matrix Mu…
We present and experimentally evaluate using transfer learning to address experimental data scarcity when training neural network (NN) models for Mach-Zehnder interferometer mesh-based optical matrix multipliers. Our approach involves…
Deployment of modern TinyML tasks on small battery-constrained IoT devices requires high computational energy efficiency. Analog In-Memory Computing (IMC) using non-volatile memory (NVM) promises major efficiency improvements in deep neural…
Traditionally, measuring the center-of-mass (c.m.) velocity of an atomic ensemble relies on measuring the Doppler shift of the absorption spectrum of single atoms in the ensemble. Mapping out the velocity distribution of the ensemble is…
Time-domain nonvolatile in-memory computing (TD-nvIMC) offers a promising pathway to reduce data movement and improve energy efficiency by encoding computation in delay rather than voltage or current. This work presents a fully integrated…
Silicon photonics has emerged as the leading candidate for implementing ultralow power wavelength division multiplexed communication networks in high-performance computers, yet current components (lasers, modulators, filters, and detectors)…
Memristive Processing In-Memory (PIM) is one of the promising techniques for overcoming the Von-Neumann bottleneck. Reduction of data transfer between processor and memory and data processing by memristors in data-intensive applications…
An analog neural network computing engine based on CMOS-compatible charge-trap transistor (CTT) is proposed in this paper. CTT devices are used as analog multipliers. Compared to digital multipliers, CTT-based analog multiplier shows…
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…
The performance of a silicon photomultiplier has been assessed at low temperature in order to evaluate its suitability as a scintillation readout device in liquid argon particle physics detectors. The gain, measured as 2.1E6 for a constant…
We investigate the thermal transport properties of a temperature-biased Josephson tunnel junction composed of two different superconductors. We show that this simple system can provide a large negative differential thermal conductance…
Analog computing is attractive compared to digital computing due to its potential for achieving higher computational density and higher energy efficiency. However, unlike digital circuits, conventional analog computing circuits cannot be…
Superconductor electronics (SCE) is a promising complementary and beyond CMOS technology. However, despite its practical benefits, the realization of SCE logic faces a significant challenge due to the absence of dense and scalable…
Matrix multiplications between asymmetric bit-width operands, especially between 8- and 4-bit operands are likely to become a fundamental kernel of many important workloads including neural networks and machine learning. While existing SIMD…
The advancement of large language models has led to models with billions of parameters, significantly increasing memory and compute demands. Serving such models on conventional hardware is challenging due to limited DRAM capacity and high…
A photonic heat amplifier (PHA) designed for cryogenic operations is introduced and analyzed. This device comprises two variable-range-hopping reservoirs connected by lossless lines, which allow them to exchange heat through photonic modes.…
This paper presents a differential 10-bit 2 MS/s successive approximation register (SAR) analog-to-digital converter (ADC) with a precision-improvement technique. The proposed method breaks the direct tradeoff between the capacitive…
Analog In-Memory Compute (AIMC) can improve the energy efficiency of Deep Learning by orders of magnitude. Yet analog-domain device and circuit non-idealities -- within the analog ``Tiles'' performing Matrix-Vector Multiply (MVM) operations…
A novel inline data compression method is presented for single-precision vectors in three dimensions. The primary application of the method is for accelerating computational physics calculations where the throughput is bound by memory…
Flash memory-based processing-in-memory (flash-based PIM) offers high storage capacity and computational efficiency but faces significant reliability challenges due to noise in high-density multi-level cell (MLC) flash memories. Existing…
The rapid growth of deep neural network (DNN) workloads has significantly increased the demand for large-capacity on-chip SRAM in machine learning (ML) applications, with SRAM arrays now occupying a substantial fraction of the total die…