Optimal order of accuracy of search algorithms in stochastic optimization
Consider min\(\{\) f(x): \(x\in Q\subset R^ N\}\) with f: \(R^ N\to R^ 1\). Assume the unique existence of a solution. f is unknown and observations \(y_ n=f(x_ n)+\xi_ n\) with random errors \(\xi_ n\) are possible. The following random search algorithm is discussed: \[ z_ 0=b\quad (given),\quad x_ n=z_{n-1}+h_ n\psi_ n\quad (h_ n>0), \] \[ z_ n=\Pr_{Q_ n}(z_{n-1}-an^{-1}h_ n^{-1}y_ nK(\psi_ n))\quad (a>0), \] where \(\psi_ 1,\psi_ 2,..\). are independent random variables uniformly distributed on [-1/2, \(1/2]^ N\), \(Q_ n=Q\) convex, compact, Pr is the projection operator, K(t) is a vector function with components of the form \[ K_ j(t)=K_ 0(t_ j)\prod^{N}_{m\neq j}\tilde K(t_ m)\quad (j=1,...,N). \] Under some regularity conditions on f it is proved that the convergence order is not higher than \(O(n^{-(\beta -1)/\beta})\) where it is assumed that \[ | f(z)-\sum_{| m| \leq 1}\quad (m!)^{-1}D^ mf(x)(z-x)^ m| \leq L| z-x|^{\beta}. \]
- scientific article; zbMATH DE number 4080680
- scientific article; zbMATH DE number 3862941
- Speed of convergence as a function of given accuracy for random search methods
- On Linear Convergence of a Class of Random Search Algorithms
- On the rate of convergence of one inhomogeneous Markov algorithm of search for extremum
- The stochastic approximation method for the estimation of a multivariate probability density
- Statistical inferences for termination of Markov type random search algorithms
- Optimum parameters and nonasymptotic bounds on the rate of convergence of stochastic algorithms in criterial optimization problems
- Recursion algorithms for search optimization in relative noise. I: Maximal possibilities
- Recurrent search optimization algorithms in the presence of relative noise. II: Optimal search procedures
- Stochastic approximation of global minimum points
- Accelerated randomized stochastic optimization.
- Improved exploitation of higher order smoothness in derivative-free optimization
- Stochastic approximation algorithm with randomization at the input for unsupervised parameters estimation of Gaussian mixture model with sparse parameters
- Bayesian mode and maximum estimation and accelerated rates of contraction
- A companion for the Kiefer-Wolfowitz-Blum stochastic approximation algorithm
- A randomized stochastic optimization algorithm: its estimation accuracy
- Algorithm for stochastic approximation with trial input perturbation in the nonstationary problem of optimization
- Stopping rules for optimization algorithms based on stochastic approximation
- scientific article; zbMATH DE number 3862941 (Why is no real title available?)
- scientific article; zbMATH DE number 3905762 (Why is no real title available?)
- scientific article; zbMATH DE number 3920219 (Why is no real title available?)
- scientific article; zbMATH DE number 3936951 (Why is no real title available?)
- scientific article; zbMATH DE number 4045480 (Why is no real title available?)
- scientific article; zbMATH DE number 4080680 (Why is no real title available?)
- scientific article; zbMATH DE number 1187210 (Why is no real title available?)
- An accelerated method for derivative-free smooth stochastic convex optimization
- AN OPTIMAL INVESTIVATION IN TWO STAGE SEARCH WITH RECOGNITION ERRORS
- Non-smooth setting of stochastic decentralized convex optimization problem over time-varying graphs
- Accelerated gradient methods with absolute and relative noise in the gradient
- Re-thinking high-dimensional mathematical statistics. Abstracts from the workshop held May 15--21, 2022
- Estimation and inference for minimizer and minimum of convex functions: optimality, adaptivity and uncertainty principles
- Polyak's method based on the stochastic Lyapunov function for justifying the consistency of estimates produced by a stochastic approximation search algorithm under an unknown-but-bounded noise
- Adaptive control using stochastic approach for unknown but bounded disturbances and its application in balancing control
- Small errors in random zeroth-order optimization are imaginary
- Stochastic adversarial noise in the ``black box optimization problem
- Online Statistical Inference for Stochastic Optimization via Kiefer-Wolfowitz Methods
- The ``black-box optimization problem: zero-order accelerated stochastic method via kernel approximation
- Optimal two-stage procedures for estimating location and size of the maximum of a multivariate regression function
- Stochastic approximation search algorithms with randomization at the input
- Criterion algorithms of stochastic optimization
This page was built for publication: Optimal order of accuracy of search algorithms in stochastic optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q749450)
