NIST · ИТ · ca455f7362ff876de868
Photonic Online Learning
Training in machine learning necessarily involves more operations than inference only, with higher precision, more memory, and added computational complexity. In hardware, many implementations side-step this issue by designing "inference-only" hardware
NIST · ИТОфициальный источник · 2024-01-09
{
"title_en": "Photonic Online Learning",
"designation": "",
"series": "",
"status": "",
"date": "2024-01-09",
"date_original": "January 9, 2024",
"url": "https://www.nist.gov/publications/photonic-online-learning",
"source": "NIST Publications",
"display_en": "",
"authors_en": "Sonia Buckley , Adam McCaughan , Bakhrom Oripov",
"abstract_excerpt_en": "Training in machine learning necessarily involves more operations than inference only, with higher precision, more memory, and added computational complexity. In hardware, many implementations side-step this issue by designing \"inference-only\" hardware",
"id": "ca455f7362ff876de868",
"sources": [
"NIST Publications"
],
"source_records": [
{
"title_en": "Photonic Online Learning",
"designation": "",
"series": "",
"status": "",
"date": "2024-01-09",
"date_original": "January 9, 2024",
"url": "https://www.nist.gov/publications/photonic-online-learning",
"source": "NIST Publications",
"display_en": "",
"authors_en": "Sonia Buckley , Adam McCaughan , Bakhrom Oripov",
"abstract_excerpt_en": "Training in machine learning necessarily involves more operations than inference only, with higher precision, more memory, and added computational complexity. In hardware, many implementations side-step this issue by designing \"inference-only\" hardware"
}
],
"translation_state": "none"
}