DeepMaker: Deep Learning Accelerator on Commercial Programmable Devices

DeepMaker aims to provide a framework to generate synthesizable accelerators of Deep Neural Networks (DNNs) that can be used for different FPGA fabrics. DeepMaker enables effective use of DNN acceleration in commercially available devices that can accelerate a wide range of applications without a need of costly FPGA reconfigurations. 

[Show all publications]

FaCT-LSTM: Fast and Compact Ternary Architecture for LSTM Recurrent Neural Networks (Jun 2022)
Najmeh Nazari , Seyed Ahmad Mirsalari , Sima Sinaei, Mostafa Salehi , Masoud Daneshtalab
IEEE Design and Test (IEEE D&T)

FastStereoNet: A Fast Neural Architecture Search for Improving the Inference of Disparity Estimation on Resource-Limited Platforms (Nov 2021)
Mohammad Loni, Ali Zoljodi, Amin Majd , Byung Hoon Ahn , Masoud Daneshtalab, Mikael Sjödin, Hadi Esmaeilzadeh
IEEE Transactions on Systems, Man, and Cybernetics: Systems (SMCS)

ELC-ECG: Efficient LSTM Cell for ECG Classification based on Quantized Architecture (May 2021)
Seyed Ahmad Mirsalari , Najmeh Nazari , Sima Sinaei, Mostafa Salehi , Masoud Daneshtalab
IEEE International Symposium on Circuits & Systems (ISCAS)

DeepMaker: Customizing the Architecture of Convolutional Neural Networks for Resource-Constrained Platforms (Dec 2020)
Mohammad Loni

MuBiNN: Multi-Level Binarized Recurrent Neural Network for EEG signal Classification (Oct 2020)
Seyed Ahmad Mirsalari , Sima Sinaei, Mostafa Salehi , Masoud Daneshtalab
IEEE International Symposium on Circuits & Systems (ISCAS)

A software implemented comprehensive soft error detection method for embedded systems (Sep 2020)
Seyyed Amir Asghari , Mohammadreza Binesh Marvasti , Masoud Daneshtalab
Elsevier journal of Microprocessors and Microsystems (MICPRO)