Zobrazeno 1 - 6
of 6
pro vyhledávání: '"MohammadHossein AskariHemmat"'
Autor:
Mohammadhossein Askarihemmat, Sean Wagner, Olexa Bilaniuk, Yassine Hariri, Yvon Savaria, Jean-Pierre David
We present a DNN accelerator that allows inference at arbitrary precision with dedicated processing elements that are configurable at the bit level. Our DNN accelerator has 8 Processing Elements controlled by a RISC-V controller with a combined 8.2 T
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9c36cbb00fb88578fb7398f50a9d5230
http://arxiv.org/abs/2301.00290
http://arxiv.org/abs/2301.00290
Autor:
Anush Sankaran, Olivier Mastropietro, Ehsan Saboori, Yasser Idris, Davis Sawyer, MohammadHossein AskariHemmat, Ghouthi Boukli Hacene
Publikováno v:
Proceedings of the AAAI Conference on Artificial Intelligence. 35:15166-15174
Designing deep learning-based solutions is becoming a race for training deeper models with a greater number of layers. While a large-size deeper model could provide competitive accuracy, it creates a lot of logistical challenges and unreasonable reso
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::b3d1f4725606883e03a9bd539564bb59
https://doi.org/10.1609/aaai.v35i17.17780
https://doi.org/10.1609/aaai.v35i17.17780
Publikováno v:
ISCAS
This paper presents a barrel RISC-V processor designed to control a deep neural network accelerator. Our design has a 5-stage pipeline data path with 8 hardware threads (harts). Each thread is executed under a strict round robin scheduler and is resp
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::154b74d57f0a4013ec516796e3ceba60
https://doi.org/10.1109/iscas51556.2021.9401617
https://doi.org/10.1109/iscas51556.2021.9401617
Publikováno v:
FCCM
Hardware accelerators are important in the post-Moore’s law era of computing. To maximize performance of such accelerators, most of the logic resources should be allocated to their execution circuits, while control mechanisms should be kept small y
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e888eea010aa98ecfe1a7a8462d72a68
https://doi.org/10.1109/fccm48280.2020.00063
https://doi.org/10.1109/fccm48280.2020.00063
Autor:
Lucas Rouhier, Sina Honari, Jean-Pierre David, Yvon Savaria, Christian S. Perone, Julien Cohen-Adad, MohammadHossein AskariHemmat
Publikováno v:
Lecture Notes in Computer Science ISBN: 9783030336417
LABELS/HAL-MICCAI/CuRIOUS@MICCAI
LABELS/HAL-MICCAI/CuRIOUS@MICCAI
Model quantization is leveraged to reduce the memory consumption and the computation time of deep neural networks. This is achieved by representing weights and activations with a lower bit resolution when compared to their high precision floating poi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::680c90a7f7a73ccac0ef3b619d79dafc
https://doi.org/10.1007/978-3-030-33642-4_13
https://doi.org/10.1007/978-3-030-33642-4_13
Publikováno v:
ICM
A new rapid prototyping methodology for embedded system application is introduced. The proposed methodology uses AutoFOCUS3(AF3), which is a model-based development tool for distributed, reactive, embedded software systems, to generate an executable
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::bfde9e02fd66c0210bf57648035a129d
https://doi.org/10.1109/icm.2014.7071849
https://doi.org/10.1109/icm.2014.7071849