Sequential design for ranking response surfaces
From MaRDI portal
Publication:5269860
Abstract: We propose and analyze sequential design methods for the problem of ranking several response surfaces. Namely, given response surfaces over a continuous input space , the aim is to efficiently find the index of the minimal response across the entire . The response surfaces are not known and have to be noisily sampled one-at-a-time. This setting is motivated by stochastic control applications and requires joint experimental design both in space and response-index dimensions. To generate sequential design heuristics we investigate stepwise uncertainty reduction approaches, as well as sampling based on posterior classification complexity. We also make connections between our continuous-input formulation and the discrete framework of pure regret in multi-armed bandits. To model the response surfaces we utilize kriging surrogates. Several numerical examples using both synthetic data and an epidemics control problem are provided to illustrate our approach and the efficacy of respective adaptive designs.
Recommendations
Cites work
- A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output from a Computer Code
- A nonstationary space-time Gaussian process model for partially converged simulations
- A probabilistic numerical method for optimal multiple switching problems in high dimension
- BART: Bayesian additive regression trees
- Buy-low and sell-high investment strategies
- Dynamic trees for learning and design
- Efficient global optimization of expensive black-box functions
- Finite-time analysis of the multiarmed bandit problem
- Gaussian processes for machine learning.
- Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit Setting
- MONTE CARLO METHODS FOR THE VALUATION OF MULTIPLE‐EXERCISE OPTIONS
- Monte Carlo algorithms for optimal stopping and statistical learning
- On posterior consistency in nonparametric regression problems
- Pure exploration in finitely-armed and continuous-armed bandits
- Sequential Bayesian inference in hidden Markov stochastic kinetic models with application to detection and response to seasonal epidemics
- Sequential design for optimal stopping problems
- Sequential design of computer experiments for the estimation of a probability of failure
- Smart Sleeping Policies for Energy Efficient Tracking in Sensor Networks
- Stochastic kriging for simulation metamodeling
- Upper-Confidence-Bound Algorithms for Active Learning in Multi-armed Bandits
- Valuation of storage at a liquefied natural gas terminal
- \(X\)-armed bandits
Cited in
(14)- Efficient sequential experimental design for surrogate modeling of nested codes
- Sparse polynomial chaos expansions: literature survey and benchmark
- Discrete-type approximations for non-Markovian optimal stopping problems. II
- Deep learning for ranking response surfaces with applications to optimal stopping problems
- Continuous multi-task Bayesian optimisation with correlation
- Moment independent uncertainty importance measure for stochastic systems based on Gaussian process
- scientific article; zbMATH DE number 795288 (Why is no real title available?)
- Technical note—Knowledge gradient for selection with covariates: Consistency and computation
- Recent developments in machine learning methods for stochastic control and games
- Large-scale local surrogate modeling of stochastic simulation experiments
- Ranking and Selection with Covariates for Personalized Decision Making
- Evaluating Gaussian process metamodels and sequential designs for noisy level set estimation
- Practical Heteroscedastic Gaussian Process Modeling for Large Simulation Experiments
- Phenomenological forecasting of disease incidence using heteroskedastic Gaussian processes: a dengue case study
This page was built for publication: Sequential design for ranking response surfaces
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5269860)
