Neural architecture codesign for fast physics applications

We develop a pipeline to streamline neural architecture codesign for physics applications to reduce the need for ML expertise when designing models for novel tasks. Our method employs neural architecture search and network compression in a two-stage approach to discover hardware efficient models. Th...

Бүрэн тодорхойлолт

Номзүйн дэлгэрэнгүй

Үндсэн зохиолчид:	Jason Weitz, Dmitri Demler, Luke McDermott, Nhan Tran, Javier Duarte
Формат:	Artigo
Хэл сонгох:	Inglês
Хэвлэсэн:	IOP Publishing 2025-01-01
Цуврал:	Machine Learning: Science and Technology
Нөхцлүүд:	neural architecture search FPGA pruning quantization high energy physics materials science
Онлайн хандалт:	https://doi.org/10.1088/2632-2153/adede1
Шошгууд:	Шошго нэмэх Шошго байхгүй, Энэхүү баримтыг шошголох эхний хүн болох!