International Conference on Computer and Knowledge Engineering

Home / 12th International Conference on Computer and Knowledge Engineering

FAST: FPGA Acceleration of Neural Networks Training

Authors :

Alireza Borhani¹ Mohammad Hossein Goharinejad² Hamid Reza Zarandi³

1- Department of Computer Engineering, Amirkabir university of technology 2- Department of Computer Engineering, Amirkabir university of technology 3- Department of Computer Engineering, Amirkabir university of technology

Keywords :

Field Programmable Gate Array،Embedded Devices،Artificial Neural Network،Machine Learning،Approximation

Abstract :

Training state-of-the-art ANNs is computationally and memory intensive. Thus, implementing the training on embedded devices with limited resources is challenging. In order to address this challenge, we propose FAST, a low-precision method to implement and optimize ANN training on FPGA. FAST first addresses the challenge of implementing the non-polynomial sigmoid activation function by presenting a solution using PNLA methods. Then, it introduces Hardware Optimized PReLU (HOPE) activation function, which is specifically devised to reduce the required resources and increase the accuracy of computations on FPGA. We evaluated FAST against the software implementations of ANNs, using training tasks available in the MNIST benchmark. The results show that FAST improves the training speed by 8.6× and reduces the required memory size by orders of magnitude. It is worthwhile to mention that the method imposes almost no degradation in training accuracy.

Papers List

List of archived papers

Density Estimation Helps Adversarial Robustness

Afsaneh Hasanebrahimi - Bahareh Kaviani Baghbaderani - Reshad Hosseini - Ahmad Kalhor

Swin-RSCBNet: A Transformer-Based Network for Skin Cancer Segmentation with Multi-Scale and Attention Modules

Benyamin Mirab Golkhatmi - Mostafa Heydari - Mahboobeh Houshmand - Seyyed Abed Hosseini

Predicting cascading failure with machine learning methods in the interdependent networks

Mohamad Hossein Maghsoodi - Mohamad Khansari

Data-Optimized Dry Rock Property Prediction Using Ensemble and Kernel-Based ML Methods

Esmael Makarian - Hassanreza Ghasemitabar - Alireza Behinrad - Mahdi Fathi - Andisheh Alimoradi - Ayub Elyasi

Real-Time Gender Recognition with a Deep Neural Network

Samad Azimi Abriz - Majid Meghdadi

MC-BioCLIPSR: A Mamba-CNN Hybrid Network with BioMedCLIP-Guided Loss for High-Resolution Brain MRI Reconstruction

Amin Kazempour - Jafar Tanha - SeyedEhsan Roshan - Mahdi Zarrin - Haniyeh Nikkhah

New Design of Efficient Reversible Quantum Saturation Adder

Negin Mashayekhi - Mohammad Reza Reshadinezhad - Shekoofeh Moghimi

FinTNet: From Tweets to Trades

Dorsa Tavakoli - Saman Haratizadeh

Object Detection on Detecting Skin Lesion using Dab-DETR

Sheida Shadman - Amirreza Rouhbakhshmeghrazi - Shayan Nalbandian - Bo Li - Shaghayegh Shadman - Malik Muhammad Owais Siddique

Recommending Popular Locations Based on Collected Trajectories

Mohammad Rabbani bidgoli - Saber Ziaei