Optimal Initialization Scale for Neural Networks With Locally Quadratic Loss Landscapes: An SGD Dynamics Perspective.

Horii H, Has S

Open source

DOI
10.1109/tnnls.2026.3721153
Published
2026 Aug 17
Container
IEEE transactions on neural networks and learning systems
Publisher
Not recorded
Open access
no

Credibility signals

limited evidence Score 43/100 under policy 1.0.0. This is a metadata assessment, not a judgment of the paper's conclusions.

Show all credibility signals

Cite this work

BibTeX

@article{allodium:10.1109/tnnls.2026.3721153,
  title = {Optimal Initialization Scale for Neural Networks With Locally Quadratic Loss Landscapes: An SGD Dynamics Perspective.},
  author = {Horii H and Has S},
  year = {2026},
  journal = {IEEE transactions on neural networks and learning systems},
  doi = {10.1109/tnnls.2026.3721153},
  url = {https://doi.org/10.1109/tnnls.2026.3721153}
}

RIS

TY  - JOUR
TI  - Optimal Initialization Scale for Neural Networks With Locally Quadratic Loss Landscapes: An SGD Dynamics Perspective.
AU  - Horii H
AU  - Has S
PY  - 2026
JO  - IEEE transactions on neural networks and learning systems
DO  - 10.1109/tnnls.2026.3721153
UR  - https://doi.org/10.1109/tnnls.2026.3721153
ER  -

APA

H, H., & S, H. (2026). Optimal Initialization Scale for Neural Networks With Locally Quadratic Loss Landscapes: An SGD Dynamics Perspective.. IEEE transactions on neural networks and learning systems. https://doi.org/10.1109/tnnls.2026.3721153

Source records