A novel backpropagation algorithm based on negated kurtosis loss for training shallow, convolutional, and deep neural networks

Engin Cemal Mengüç, Alper Emlek, Danilo P. Mandic

Open source

DOI
10.1016/j.neunet.2026.108570
Published
2026-06
Container
Neural Networks
Publisher
Elsevier BV
Open access
unknown

Credibility signals

uncertain Score 64/100 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:10.1016/j.neunet.2026.108570,
  title = {A novel backpropagation algorithm based on negated kurtosis loss for training shallow, convolutional, and deep neural networks},
  author = {Engin Cemal Mengüç and Alper Emlek and Danilo P. Mandic},
  year = {2026},
  journal = {Neural Networks},
  doi = {10.1016/j.neunet.2026.108570},
  url = {https://doi.org/10.1016/j.neunet.2026.108570}
}

RIS

TY  - JOUR
TI  - A novel backpropagation algorithm based on negated kurtosis loss for training shallow, convolutional, and deep neural networks
AU  - Engin Cemal Mengüç
AU  - Alper Emlek
AU  - Danilo P. Mandic
PY  - 2026
JO  - Neural Networks
DO  - 10.1016/j.neunet.2026.108570
UR  - https://doi.org/10.1016/j.neunet.2026.108570
ER  -

APA

Mengüç, E. C., Emlek, A., & Mandic, D. P. (2026). A novel backpropagation algorithm based on negated kurtosis loss for training shallow, convolutional, and deep neural networks. Neural Networks. https://doi.org/10.1016/j.neunet.2026.108570

Source records