General Results on the Convergence of Stochastic Algorithms
نویسندگان
چکیده
We prove, by using a deterministic approach, the convergence of stochastic algorithms of the most general form, under necessary conditions on the input noise, and reasonable and veriiable conditions on the (non-necessarily continuous) mean eld. We also consider the case where the mean eld has a non-discrete set of stationary points. These results are also encapsulated into the familiar Markovian setup for stochastic approximation. Convergence des algorithmes stochastiques a plusieurs points stationnaires R esum e : On propose ici une approche d eterministe au probl eme de la convergence des algorithmes sto-chastiques g en eraux. Les conditions demand ees sur le bruit d'excitation sont n ecessaires et suusantes et les conditions sur le champ moyen sont raisonnables et faciles a v eriier. On donne aussi des conditions sous les-quelles la convergence vers un point a lieu quand le champ moyen poss ede une ensemble non-discret de points stationnaires. On etudie egalement les probl emes de projection ainsi que les algorithmes a dynamique marko-vienne.
منابع مشابه
On the Convergence Analysis of Gravitational Search Algorithm
Gravitational search algorithm (GSA) is one of the newest swarm based optimization algorithms, which has been inspired by the Newtonian laws of gravity and motion. GSA has empirically shown to be an efficient and robust stochastic search algorithm. Since introducing GSA a convergence analysis of this algorithm has not yet been developed. This paper introduces the first attempt to a formal conve...
متن کاملOn the Convergence Analysis of Gravitational Search Algorithm
Gravitational search algorithm (GSA) is one of the newest swarm based optimization algorithms, which has been inspired by the Newtonian laws of gravity and motion. GSA has empirically shown to be an efficient and robust stochastic search algorithm. Since introducing GSA a convergence analysis of this algorithm has not yet been developed. This paper introduces the first attempt to a formal conve...
متن کاملTraining Radial Basis Function Neural Network using Stochastic Fractal Search Algorithm to Classify Sonar Dataset
Radial Basis Function Neural Networks (RBF NNs) are one of the most applicable NNs in the classification of real targets. Despite the use of recursive methods and gradient descent for training RBF NNs, classification improper accuracy, failing to local minimum and low-convergence speed are defects of this type of network. In order to overcome these defects, heuristic and meta-heuristic algorith...
متن کاملAn Analytical Model for Predicting the Convergence Behavior of the Least Mean Mixed-Norm (LMMN) Algorithm
The Least Mean Mixed-Norm (LMMN) algorithm is a stochastic gradient-based algorithm whose objective is to minimum a combination of the cost functions of the Least Mean Square (LMS) and Least Mean Fourth (LMF) algorithms. This algorithm has inherited many properties and advantages of the LMS and LMF algorithms and mitigated their weaknesses in some ways. The main issue of the LMMN algorithm is t...
متن کاملModify the linear search formula in the BFGS method to achieve global convergence.
<span style="color: #333333; font-family: Calibri, sans-serif; font-size: 13.3333px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: justify; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; background-color: #ffffff; text-dec...
متن کاملOnline Learning and Stochastic Approximations
The convergence of online learning algorithms is analyzed using the tools of the stochastic approximation theory, and proved under very weak conditions. A general framework for online learning algorithms is first presented. This framework encompasses the most common online learning algorithms in use today, as illustrated by several examples. The stochastic approximation theory then provides gen...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1996