Skip to yearly menu bar Skip to main content


Poster

Bounds for the smallest eigenvalue of the NTK for arbitrary spherical data of arbitrary dimension

Kedar Karhadkar · Michael Murray · Guido Montufar

East Exhibit Hall A-C #2310
[ ]
[ Paper [ Poster [ OpenReview
Fri 13 Dec 11 a.m. PST — 2 p.m. PST

Abstract: Bounds on the smallest eigenvalue of the neural tangent kernel (NTK) are a key ingredient in the analysis of neural network optimization and memorization. However, existing results require distributional assumptions on the data and are limited to a high-dimensional setting, where the input dimension d0 scales at least logarithmically in the number of samples n. In this work we remove both of these requirements and instead provide bounds in terms of a measure of distance between data points: notably these bounds hold with high probability even when d0 is held constant versus n. We prove our results through a novel application of the hemisphere transform.

Chat is not available.