Poster
in
Workshop: NeurIPS 2023 Workshop: Machine Learning and the Physical Sciences
19 Parameters Is All You Need: Tiny Neural Networks for Particle Physics
Alexander Bogatskiy · Timothy Hoffman · Jan Offermann
Abstract:
As particle accelerators increase their collision rates, and machine learning solutionsprove their reliability, the need for lightweight and fast neural network hardwareimplementations grows for low-latency tasks such as triggering. We examine thepotential of one recent Lorentz- and permutation-symmetric architecture, PELICAN,and present its instances with as few as 19 trainable parameters that outperformgeneric architectures with tens of thousands of parameters when compared on thebinary classification task of top quark jet tagging.
Chat is not available.