Efficient hardware architecture of softmax layer in deep neural network

R Hu, B Tian, S Yin, S Wei - 2018 IEEE 23rd International …, 2018 - ieeexplore.ieee.org
Deep neural network (DNN), as a very important machine learning technique in
classification and detection tasks for images, video, speech as wellas audio, has recently …

Efficient precision-adjustable architecture for softmax function in deep learning

D Zhu, S Lu, M Wang, J Lin… - IEEE Transactions on …, 2020 - ieeexplore.ieee.org
The softmax function has been widely used in deep neural networks (DNNs), and studies on
efficient hardware accelerators for DNN have also attracted tremendous attention. However …

FPGA-based implementation of deep neural network using stochastic computing

M Nobari, H Jahanirad - Applied Soft Computing, 2023 - Elsevier
A serious challenge in artificial real-time applications is the hardware implementation of
deep neural networks (DNN). Among various methods, stochastic computing (SC)-based …

Base-2 softmax function: Suitability for training and efficient hardware implementation

Y Zhang, Y Zhang, L Peng, L Quan… - … on Circuits and …, 2022 - ieeexplore.ieee.org
The softmax function is widely used in deep neural networks (DNNs), its hardware
performance plays an important role in the training and inference of DNN accelerators …

[HTML][HTML] Design of infinite impulse response filters based on multi-objective particle Swarm optimization

TJ Su, QY Zhuang, WH Lin, YC Hung, WR Yang… - Signals, 2024 - mdpi.com
The goal of this study is to explore the effectiveness of applying multi-objective particle
swarm optimization (MOPSO) algorithms in the design of infinite impulse response (IIR) …

High-accuracy stochastic computing-based fir filter design

KJ Ahmed, B Yuan, MJ Lee - 2018 IEEE International …, 2018 - ieeexplore.ieee.org
In this paper we propose a novel Stochastic Computing (SC) Finite Impulse Response (FIR)
filter design to improve the overall accuracy. Since SC is generated probabilistically, it incurs …

All-digital time-mode direct-form all-pole biquadratic filter realization

M Abdelfattah, GW Roberts - IEEE Transactions on Circuits and …, 2016 - ieeexplore.ieee.org
In this brief, the design of time-mode signal processing (TMSP) circuits having an all-digital
advantage is introduced and experimental data covering all necessary aspects of operation …

Low-cost approximate constant coefficient hybrid binary-unary multiplier for DSP applications

SR Faraji, P Abillama… - 2020 IEEE 28th Annual …, 2020 - ieeexplore.ieee.org
Multipliers are used in virtually all Digital Signal Processing (DSP) applications, such as
image and video processing. Multiplier efficiency has a direct impact on the overall …

Hardware implementation of infinite impulse response anti‐notch filter for exon region identification in eukaryotic genes

V Pathak, S Jagannath Nanda… - … Journal of Circuit …, 2020 - Wiley Online Library
The discrimination of exons from introns in the DNA sequence of a eukaryotic gene is
important to understand the functionality of protein formation inside a living organism …

Optimization of softmax layer in deep neural network using integral stochastic computation

R Hu, B Tian, S Yin, S Wei - Journal of Low Power Electronics, 2018 - ingentaconnect.com
Deep neural network (DNN), as a very important machine learning technique in
classification and detection tasks for images, video, speech as well as audio, has recently …