Parameter Tuning in Pointwise Adaptation Using a Propagation Approach
نویسنده
چکیده
This paper discusses the problem of adaptive estimation of a univariate object like the value of a regression function at a given point or a linear functional in a linear inverse problem. We consider an adaptive procedure originated from Lepski [Theory Probab. Appl. 35 (1990) 454–466.] that selects in a data-driven way one estimate out of a given class of estimates ordered by their variability. A serious problem with using this and similar procedures is the choice of some tuning parameters like thresholds. Numerical results show that the theoretically recommended proposals appear to be too conservative and lead to a strong oversmoothing effect. A careful choice of the parameters of the procedure is extremely important for getting the reasonable quality of estimation. The main contribution of this paper is the new approach for choosing the parameters of the procedure by providing the prescribed behavior of the resulting estimate in the simple parametric situation. We establish a non-asymptotical " oracle " bound, which shows that the estimation risk is, up to a logarithmic multiplier, equal to the risk of the " oracle " estimate that is optimally selected from the given family. A numerical study demonstrates a good performance of the resulting procedure in a number of simulated examples.
منابع مشابه
Efficient and Robust Parameter Tuning for Heuristic Algorithms
The main advantage of heuristic or metaheuristic algorithms compared to exact optimization methods is their ability in handling large-scale instances within a reasonable time, albeit at the expense of losing a guarantee for achieving the optimal solution. Therefore, metaheuristic techniques are appropriate choices for solving NP-hard problems to near optimality. Since the parameters of heuristi...
متن کاملA Unified IMC based PI/PID Controller Tuning Approach for Time Delay Processes
This paper proposes a new PI/PID controller tuning method within filtered Smith predictor (FSP) configuration in order to deal with various types of time delay processes including stable, unstable and integrating delay dominant and slow dynamic processes. The proposed PI/PID controller is designed based on the IMC principle and is tuned using a new constraint and without requiring any approxima...
متن کاملAdaptive Tuning of Model Predictive Control Parameters based on Analytical Results
In dealing with model predictive controllers (MPC), controller tuning is a key design step. Various tuning methods are proposed in the literature which can be categorized as heuristic, numerical and analytical methods. Among the available tuning methods, analytical approaches are more interesting and useful. This paper is based on a proposed analytical MPC tuning approach for plants can be appr...
متن کاملOptimal Control of Light Propagation Governed by Eikonal Equation within Inhomogeneous Media Using Computational Adjoint Approach
A mathematical model is presented in the present study to control the light propagation in an inhomogeneous media. The method is based on the identification of the optimal materials distribution in the media such that the trajectories of light rays follow the desired path. The problem is formulated as a distributed parameter identification problem and it is solved by a numerical met...
متن کاملAdaptive Line Enhancement Using a Parallel IIR Filter with A Step-By-step Algorithm
A step-by-step algorithm for enhancement of periodic signals that are highly corrupted by additive uncorrelated white gausian noise is proposed. In each adaptation step a new parallel second-order section is added to the previous filters. Every section has only one adjustable parameter, i.e., the center frequency of the self-tuning filter. The bandwidth and the convergence factor of each secti...
متن کامل