ArXiv.org · 2026 · 79 citations · 14 references
Natural SciencesComputer ScienceAlgorithmsOptimizationStochastic ProcessesApproximation AlgorithmsConvex OptimizationGlobal OptimizationOptimization TheoryOptimization Under UncertaintyStochastic OptimizationBandit AlgorithmsFinite TimeNoisy OptimizationStochastic BanditsUpper Bounds
We study the problem of global maximization of a function f given a finite number of evaluations perturbed by noise. We consider a very weak assumption on the function, namely that it is locally smooth (in some precise sense) with respect to some semi-metric, around one of its global maxima. Compared to previous works on bandits in general spaces (Kleinberg et al., 2008; Bubeck et al., 2011a) our algorithm does not require the knowledge of this semi-metric. Our algorithm, StoSOO, follows an optimistic strategy to iteratively construct upper confidence bounds over the hierarchical partitions of the function domain to decide which point to sample next. A finite-time analysis of StoSOO shows that it performs almost as well as the best specifically-tuned algorithms even though the local smoothness of the function is not known.
14
Nonconvex Optimization and Its Applications
Pãnos M. Pardalos, Shashi Kant Mishra, Shouyang Wang et al. · 2008 · 533 citations