ML Anthology

A Stochastic Momentum Accelerated Quasi-Newton Method for Neural Networks (Student Abstract)

S. Indrapriyadarsini, Shahrzad Mahboubi, Hiroshi Ninomiya, Takeshi Kamio, Hideki Asai
AAAI 2022 pp. 12973-12974
doi:10.1609/AAAI.V36I11.21623 /aaai/2022/indrapriyadarsini2022aaai-stochastic/

Abstract

Incorporating curvature information in stochastic methods has been a challenging task. This paper proposes a momentum accelerated BFGS quasi-Newton method in both its full and limited memory forms, for solving stochastic large scale non-convex optimization problems in neural networks (NN).

PDF AAAI Semantic Scholar

Cite

Text

Indrapriyadarsini et al. "A Stochastic Momentum Accelerated Quasi-Newton Method for Neural Networks (Student Abstract)." AAAI Conference on Artificial Intelligence, 2022. doi:10.1609/AAAI.V36I11.21623

Markdown

[Indrapriyadarsini et al. "A Stochastic Momentum Accelerated Quasi-Newton Method for Neural Networks (Student Abstract)." AAAI Conference on Artificial Intelligence, 2022.](https://mlanthology.org/aaai/2022/indrapriyadarsini2022aaai-stochastic/) doi:10.1609/AAAI.V36I11.21623

BibTeX

@inproceedings{indrapriyadarsini2022aaai-stochastic,
  title     = {{A Stochastic Momentum Accelerated Quasi-Newton Method for Neural Networks (Student Abstract)}},
  author    = {Indrapriyadarsini, S. and Mahboubi, Shahrzad and Ninomiya, Hiroshi and Kamio, Takeshi and Asai, Hideki},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2022},
  pages     = {12973-12974},
  doi       = {10.1609/AAAI.V36I11.21623},
  url       = {https://mlanthology.org/aaai/2022/indrapriyadarsini2022aaai-stochastic/}
}