We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Adaptive Learning Rate via Covariance Matrix Based Preconditioning for Deep Neural Networks

Abstract: Adaptive learning rate algorithms such as RMSProp are widely used for training deep neural networks. RMSProp offers efficient training since it uses first order gradients to approximate Hessian-based preconditioning. However, since the first order gradients include noise caused by stochastic optimization, the approximation may be inaccurate. In this paper, we propose a novel adaptive learning rate algorithm called SDProp. Its key idea is effective handling of the noise by preconditioning based on covariance matrix. For various neural networks, our approach is more efficient and effective than RMSProp and its variant.
Comments: Accepted at IJCAI 2017
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
DOI: 10.24963/ijcai.2017/267
Cite as: arXiv:1605.09593 [cs.LG]
  (or arXiv:1605.09593v2 [cs.LG] for this version)

Submission history

From: Yasutoshi Ida [view email]
[v1] Tue, 31 May 2016 12:11:51 GMT (29kb)
[v2] Thu, 28 Sep 2017 10:24:50 GMT (48kb,D)

Link back to: arXiv, form interface, contact.