[HTML][HTML] Noise tailoring, noise annealing, and external perturbation injection strategies in memristive Hopfield neural networks
JG Fehérvári, Z Balogh, TN Török… - APL Machine Learning, 2024 - pubs.aip.org
The commercial introduction of a novel electronic device is often preceded by a lengthy
material optimization phase devoted to the suppression of device noise as much as …
material optimization phase devoted to the suppression of device noise as much as …
Energy Efficient Memristive Transiently Chaotic Neural Network for Combinatorial Optimization
H Bao, P Ren, K Xu, L Yang, H Zhou… - … on Circuits and …, 2024 - ieeexplore.ieee.org
The utilization of memristive analog-digital mixed in-memory computing has significantly
tackled the issues of massive computing resources and time delays in solving combinatorial …
tackled the issues of massive computing resources and time delays in solving combinatorial …
A Unified Framework for Combinatorial Optimization Based on Graph Neural Networks
Graph neural networks (GNNs) have emerged as a powerful tool for solving combinatorial
optimization problems (COPs), exhibiting state-of-the-art performance in both graph …
optimization problems (COPs), exhibiting state-of-the-art performance in both graph …
Computing High-Degree Polynomial Gradients in Memory
T Bhattacharya, GH Hutchinson, G Pedretti… - arXiv preprint arXiv …, 2024 - arxiv.org
Specialized function gradient computing hardware could greatly improve the performance of
state-of-the-art optimization algorithms, eg, based on gradient descent or conjugate gradient …
state-of-the-art optimization algorithms, eg, based on gradient descent or conjugate gradient …
Energy Efficient Knapsack Optimization Using Probabilistic Memristor Crossbars
J Li, S Kumar, S Yi - arXiv preprint arXiv:2407.04332, 2024 - arxiv.org
Constrained optimization underlies crucial societal problems (for instance, stock trading and
bandwidth allocation), but is often computationally hard (complexity grows exponentially …
bandwidth allocation), but is often computationally hard (complexity grows exponentially …
Device-Architecture Co-optimization for RRAM-based In-memory Computing
In the pursuit of high energy-efficiency computing solutions, RRAM-based in-memory
computing (IMC) technology has emerged as a promising contender, offering notable …
computing (IMC) technology has emerged as a promising contender, offering notable …
The Role of Tunneling Oxide in the Low Frequency Noise of Multi-level Silicon Nitride ReRAMs
N Vasileiadis, A Mavropoulis… - … on Noise and …, 2023 - ieeexplore.ieee.org
This research explores the characteristics of two CMOS-compatible RRAM cells utilizing
silicon nitride as the switching material. By employing SET/RESET pulse sequences, the …
silicon nitride as the switching material. By employing SET/RESET pulse sequences, the …