activity
20172026
most citedCollapse of Deep and Narrow Neural Nets

282 citations · 571 across the 9 of their papers we have counts for

collaborators

11 papers

math.NT2026

Interlacing for zeros of the Serre derivative of Eisenstein series

Maggie Bohanek, Owen McGinty, Erick Ross +2

In 1970, Rankin and Swinnerton-Dyer showed that the non-elliptic zeros of Eisenstein series in the fundamental domain all lie on the lower arc $\{ e^{iθ}: \fracπ{2} < θ< \fra…

math.NT2026

Explicit generators of the space of modular forms

Tianyu Ni, Ashley Song, Yanhui Su +2

Let be the space of cusp forms of weight and level one, and let denote its dual space. In this paper, we find explicit spanning subsets of consisting o…

math.CO2020

Graphs on surfaces with positive Forman curvature or corner curvature

Yohji Akama, Bobo Hua, Yanhui Su +1

On one hand, we study the class of graphs on surfaces, satisfying tessellation properties, with positive Forman curvature on each edge. Via medial graphs, we provide a new proof fo…

math.CO2019★ 2 cited

A curvature notion for planar graphs stable under planar duality

Yohji Akama, Bobo Hua, Yanhui Su +1

Woess \cite{Woess98} introduced a curvature notion on the set of edges of a planar graph, called -curvature in our paper, which is stable under the planar duality. We study geom…

stat.ML2019★ 282 cited

Dying ReLU and Initialization: Theory and Numerical Examples

Lu Lu, Yeonjong Shin, Yanhui Su +1

The dying ReLU refers to the problem when ReLU neurons become inactive and only output 0 for any input. There are many empirical and heuristic explanations of why ReLU neurons die.…

stat.ML2018★ 282 cited

Collapse of Deep and Narrow Neural Nets

Lu Lu, Yanhui Su, George Em Karniadakis

Recent theoretical work has demonstrated that deep neural networks have superior performance over shallow networks, but their training is more difficult, e.g., they suffer from the…