It is well known that the stochastic optimization problem can be regarded as one of the most hard problems since, in most of the cases, the values off f and its gradient are often not easily to be solved, or the F(., ...
详细信息
It is well known that the stochastic optimization problem can be regarded as one of the most hard problems since, in most of the cases, the values off f and its gradient are often not easily to be solved, or the F(., xi) is normally not given clearly and (or) the distribution function P is equivocal. Then an effective optimization algorithm is successfully designed and used to solve this problem that is an interesting work. This paper designs stochastic bigger subspacealgorithms for solving nonconvex stochastic optimization problems. A general framework for such algorithm is presented for convergence analysis, where the so-called the sufficient descent property, the trust region feature, and the global convergence of the stationary points are proved under the suitable conditions. In the worst-case, we will turn out that the complexity is competitive under a given accuracy parameter. We will proved that the SFO-calls complexity of the presented algorithm with diminishing steplength is O(epsilon( -1/1-beta)) and the SFO-calls complexity of the given algorithm with random constant steplength is O(epsilon(-2)) respectively, where beta is an element of (0.5, 1) and epsilon is accuracy and the needed conditions are weaker than the quasi-Newton methods and the normal conjugate gradient algorithms. The detail algorithm framework with variance reduction is also proposed for experiments and the nonconvex binary classification problem is done to demonstrate the performance of the given algorithm.
作者：
Miniar Attig[a]Maher Abdelghani[b]Nabil ben Kahla[c][*][a] Ecole Polytechnique de Tunisie
University of Carthage Applied Mechanics and Systems Research Laboratory Tunis Tunisia[b] Higher Institute of Applied Sciences and Technologies of Sousse University of Sousse Sousse Tunisia[c] King Khalid University Civil Engineering Department Abha Saudi Arabia
Tensegrity systems are a special class of spatial reticulated structures that are composed of struts in compression and cables in tension. In this paper, the performance of stochastic subspace algorithms for modal ide...
详细信息
Tensegrity systems are a special class of spatial reticulated structures that are composed of struts in compression and cables in tension. In this paper, the performance of stochastic subspace algorithms for modal identification of complex tensegrity systems is investigated. A sub-class algorithm of the stochasticsubspace Identification family: the Balanced Realization algorithm is investigated for modal identification of a tripod simplex structure and a Geiger dome. The presented algorithm is combined with a stabilization diagram with combined criteria (frequency, damping and mode shapes). It is shown that although the studied structures present closely spaced modes, the Balanced Realization algorithm performs well and guarantees separation between closely-spaced natural frequencies. Modal identification results are validated through comparisons of the correlations (empirical vs. model based) showing effectiveness of the proposed methodology.
暂无评论