1992 AccelerationofStochasticApproxi

Revision as of 02:34, 24 June 2021 by Gmelli (talk | contribs) (Text replacement - "==Notes==" to "== Notes ==")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Subject Headings: Stochastic Gradient Descent Algorithm.

Notes

Cited By

Quotes

Author Keywords

Abstract

A new recursive algorithm of stochastic approximation type with the averaging of trajectories is investigated. Convergence with probability one is proved for a variety of classical optimization and identification problems. It is also demonstrated for these problems that the proposed algorithm achieves the highest possible rate of convergence.

1. Introduction.

The methods of stochastic approximation originate in the works [29], [12] and are currently well studied [5], [21], [14], [16], [40].

References

  • [12] E. KIEFER AND J. WOLFOVITZ, Stochastic estimation of the maximum of a regression function, Ann. Math. Statist., 23 (1952), pp. 462-466.
  • [29] H. ROBBINS AND S. MONROE, A stochastic-approximation method, Ann. Math. Statist., 22 (1951), pp. 400-407.
  • …,


 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
1992 AccelerationofStochasticApproxiBoris T. Polyak
Anatoli B. Juditsky
Acceleration of Stochastic Approximation by Averaging10.1137/0330046