Effects of Data Geometry in Early Deep Learning
Published in NeurIPS 2022, 2022
We provide a new theory for understanding the capacity of neural networks in light of the manifold hypothesis
Recommended citation: Saket Tiwari, & George Konidaris. (2022). "Effects of Data Geometry in Early Deep Learning." NeurIPS 2022 https://openreview.net/pdf?id=QYQH9w9Z8bO