Optimal learning for sequential sampling with non-parametric beliefs
The authors consider the problem of maximizing an unknown function over a finite set of possible alternatives. They propose a sequential learning policy for ranking and selection problems, using a non-parametric procedure for estimating the value of a policy. Their estimation approach aggregates over a set of kernel functions in order to achieve a more consistent estimator. The final estimate uses a weighting scheme with the inverse \ mean square errors of the kernel estimators as weights. This weighting scheme is shown to be optimal under the independent kernel estimators. For choosing the measurement, the authors employ the knowledge gradient policy that relies on predictive distributions to calculate the optimal sampling point. The method allows a setting where the beliefs are expected to be correlated but the correlation structure is unknown beforehand. Moreover, the proposed policy is shown to be asymptotically optimal.
- Optimal Learning for Stochastic Optimization with Nonlinear Parametric Belief Models
- Sequential sampling techniques for algorithmic learning theory
- scientific article; zbMATH DE number 2080438
- Optimal learning for nonlinear parametric belief models over multidimensional continuous spaces
- Optimal online learning for nonlinear belief models using discrete priors
- Sequentially adaptive Bayesian learning algorithms for inference and optimization
- Learning to optimize via posterior sampling
- scientific article; zbMATH DE number 6276176
- scientific article; zbMATH DE number 1329952
- A decision-theoretic generalization of on-line learning and an application to boosting
- A Knowledge-Gradient Policy for Sequential Information Collection
- A Stochastic Approximation Method
- An informational approach to the global optimization of expensive-to-evaluate functions
- Approximate Dynamic Programming
- Bayesian look ahead one-stage sampling allocations for selection of the best population
- Functional aggregation for nonparametric regression.
- Global optimization of stochastic black-box systems via sequential kriging meta-models
- Hierarchical knowledge gradient for sequential sampling
- scientific article; zbMATH DE number 991833 (Why is no real title available?)
- scientific article; zbMATH DE number 3126067 (Why is no real title available?)
- scientific article; zbMATH DE number 48302 (Why is no real title available?)
- scientific article; zbMATH DE number 3474804 (Why is no real title available?)
- scientific article; zbMATH DE number 3638998 (Why is no real title available?)
- scientific article; zbMATH DE number 765034 (Why is no real title available?)
- scientific article; zbMATH DE number 795288 (Why is no real title available?)
- Introduction to Stochastic Search and Optimization
- Monotone approximation of decision problems
- Nonparametric and semiparametric models.
- Sequential Procedures for Aggregating Arbitrary Estimators of a Conditional Mean
- The Continuum-Armed Bandit Problem
- The knowledge gradient algorithm for a general class of online learning problems
- The knowledge-gradient algorithm for sequencing experiments in drug discovery
- The knowledge-gradient policy for correlated normal beliefs
- Value function approximation using multiple aggregation for multiattribute resource management
- Widely Convergent Method for Finding Multiple Solutions of Simultaneous Nonlinear Equations
- A unified framework for stochastic optimization
- Optimal learning with non-Gaussian rewards
- Sequential sampling to myopically maximize the expected value of information
- Optimal learning in linear regression with combinatorial feature selection
- Optimal learning with local nonlinear parametric models over continuous designs
- Optimal learning for nonlinear parametric belief models over multidimensional continuous spaces
- Optimal Learning for Stochastic Optimization with Nonlinear Parametric Belief Models
- Finite-time analysis for the knowledge-gradient policy
- Optimal online learning for nonlinear belief models using discrete priors
- Contextual ranking and selection with Gaussian processes and optimal computing budget allocation
- Optimal learning with a local parametric belief model
This page was built for publication: Optimal learning for sequential sampling with non-parametric beliefs
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q742143)
