Closed-Loop Generative Selection: Convergence, Memory, and Noisy Oracles

DOI10.48550/arXiv.2607.22211MaRDI QIDQ7284509zbMATH OpenarXivFDO

Authors Konstantin Fackeldey, Christof Schütte

Publication date 24 July 2026

Published in arXiv

Tldr We develop a self-contained theory for closed-loop generative selection algorithms, which we apply to analyze their convergence and expected running time, memory usage, noisy oracles, multi-objective search, and optimal evaluation strategies.

Full work available at URL https://doi.org/10.48550/arXiv.2607.22211

Mathematics Subject Classification ID

Sequential statistical design (62L05) Markov chains (discrete-time Markov processes on discrete state spaces) (60J10) Analysis of algorithms and problem complexity (68Q25) Randomized algorithms (68W20) Nonconvex programming, global optimization (90C26) Markov and semi-Markov decision processes (90C40)

arXiv classification

cs.NEmath.NA

Cites work

• Action elimination and stopping conditions for the multi-armed bandit and reinforcement learning problems
• Challenging the empirical mean and empirical variance: a deviation study
• Drift analysis and average time complexity of evolutionary algorithms
• Efficient optimisation of noisy fitness functions with population-based evolutionary algorithms
• Fitness levels with tail bounds for the analysis of randomized search heuristics
• Lower bounds on the run time of the univariate marginal distribution algorithm on OneMax
• Markov chains and stochastic stability
• Mean estimation and regression under heavy-tailed distributions: A survey
• Multi-fidelity optimization via surrogate modelling
• Multiplicative drift analysis
• On the complexity of best-arm identification in multi-armed bandit models
• Prediction, Learning, and Games
• Robustness of populations in stochastic environments
• Upper bounds on the running time of the univariate marginal distribution algorithm on OneMax

This page was built for publication: Closed-Loop Generative Selection: Convergence, Memory, and Noisy Oracles

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7284509)