ML Anthology

Two Timescale Analysis of the Alopex Algorithm for Optimization

P. S. Sastry, M. Magesh, K. P. Unnikrishnan
NeCo 2002 pp. 2729-2750
doi:10.1162/089976602760408044 /neco/2002/sastry2002neco-two/

Abstract

Alopex is a correlation-based gradient-free optimization technique useful in many learning problems. However, there are no analytical results on the asymptotic behavior of this algorithm. This article presents a new version of Alopex that can be analyzed using techniques of two timescale stochastic approximation method. It is shown that the algorithm asymptotically behaves like a gradient-descent method, though it does not need (or estimate) any gradient information. It is also shown, through simulations, that the algorithm is quite effective.

NeCo Semantic Scholar

Cite

Text

Sastry et al. "Two Timescale Analysis of the Alopex Algorithm for Optimization." Neural Computation, 2002. doi:10.1162/089976602760408044

Markdown

[Sastry et al. "Two Timescale Analysis of the Alopex Algorithm for Optimization." Neural Computation, 2002.](https://mlanthology.org/neco/2002/sastry2002neco-two/) doi:10.1162/089976602760408044

BibTeX

@article{sastry2002neco-two,
  title     = {{Two Timescale Analysis of the Alopex Algorithm for Optimization}},
  author    = {Sastry, P. S. and Magesh, M. and Unnikrishnan, K. P.},
  journal   = {Neural Computation},
  year      = {2002},
  pages     = {2729-2750},
  doi       = {10.1162/089976602760408044},
  volume    = {14},
  url       = {https://mlanthology.org/neco/2002/sastry2002neco-two/}
}