Sharp Closure Thresholds for Two-Hidden-Layer ReLU Networks
Preprint, Version 1. This manuscript has not been peer reviewed. This deposit establishes a public, citable, version-specific scholarly record of the manuscript and its accompanying source and verification materials. No journal acceptance or endorsement is implied. The SHA-256 checksums of the deposited manuscript and source/verification archive are: EBDB49404F295CD685C01DB5C2C821090DF45C96AA1F88332EBEE317A77B2461 sharp_closure_thresholds_two_hidden_layer_relu_networks_v1.pdf 215504A19CCB138A3C0635FB0FCDF95F0D0FC340A246923F6534FF09EF014928 supplementary_verification.zip The same checksums are provided in the accompanying SHA256SUMS.txt file.
Authors
- Shunqi Lu (ORCID: https://orcid.org/0009-0005-8251-2070)
Institutions
- Artificial Intelligence in Medicine (Canada) (CA)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-09-13
- DOI
- https://doi.org/10.5281/zenodo.22728502
- Primary Topic
- Advanced Graph Neural Networks
- Type
- preprint