Zero-memory-overhead clipping-based fault tolerance for LSTM deep neural networks

autor
Parchekani, Bahram
Nazari, Samira
Ahmadilivani, Mohammad Hasan
Azarpeyvand, Ali
Raik, Jaan
Ghasempouri, Tara
Daneshtalab, Masoud
vastutusandmed
Bahram Parchekani, Samira Nazari, Mohammad Hasan Ahmadilivani, Ali Azarpeyvand, Jaan Raik, Tara Ghasempouri and Masoud Daneshtalab
allikas
37th IEEE International Symposium on Defect and Fault Tolerancein VLSI and Nanotechnology Systems, Harwell, Oxfordshire, Didcot, UK, October 8th - 10th, 2024
ilmumiskoht
Piscataway, New Jersey
kirjastus/väljaandja
IEEE
ilmumisaasta
2024
leheküljed
4 p. : ill
konverentsi nimetus, aeg
37th IEEE Iternational Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems, 8-10 October, 2024
konverentsi toimumispaik
Harwell, Oxfordshire, Didcot, UK
leitav
https://doi.org/10.1109/DFT63277.2024.10753533
märksõna
tõrketaluvus
riistvara
töökindlus
tervishoid
tehisnärvivõrgud
Scopus
https://www.scopus.com/sourceid/21101123252
https://www.scopus.com/record/display.uri?eid=2-s2.0-85212407196&origin=resultslist&sort=plf-f&src=s&sot=b&sdt=b&s=ALL%28%22Zero-Memory-Overhead+Clipping-Based+Fault+Tolerance+for+LSTM+Deep+Neural+Networks%22%29&sessionSearchId=fd90d0e9983d9a88c6a21e7c6aafecbb&relpos=0
võtmesõna
fault tolerance
hardware reliability
LSTMs
neural networks
healthcare
ISSN
2576-1501
ISBN
979-835036688-4
märkused
Bibliogr.: 18 ref
teaduspublikatsioon
teaduspublikatsioon
klassifikaator
3.1
TTÜ struktuuriüksus
arvutisüsteemide instituut
keel
inglise