6–10 Nov 2023
DESY
Europe/Zurich timezone

Towards a phenomenological understanding of neural networks

9 Nov 2023, 15:15
15m
Main Auditorium (DESY)

Main Auditorium

DESY

Speaker

Samuel Tovey (University of Stuttgart)

Description

Neural networks are a powerful tool for an ever-growing list of tasks. However, their enormous complexity often complicates developing theories describing how these networks learn. In our recent work, inspired by the development of statistical mechanics, we have studied the use of collective variables to explain how neural networks learn, specifically, the von Neumann entropy and Trace of the empirical neural tangent kernel (NTK). We show that the entropy and trace of the NTK at the start of training can indicate the diversity of the training data and even predict the quality of the model after training. Further work investigates the application of these variables to understand network dynamics better, exploring optimizers for better training and the construction of better network architectures.

Primary authors

Samuel Tovey (University of Stuttgart) Dr Sven Krippendorf (LMU Munich)

Co-authors

Prof. Christian Holm (University of Stuttgart) Mr Konstantin Nikolaou (University of Stuttgart)

Presentation materials