000131433 001__ 131433
000131433 005__ 20240209155915.0
000131433 0247_ $$2doi$$a10.1109/TCYB.2018.2805695
000131433 0248_ $$2sideral$$a106896
000131433 037__ $$aART-2019-106896
000131433 041__ $$aeng
000131433 100__ $$0(orcid)0000-0002-6741-844X$$aMartinez Cantin, Ruben$$uUniversidad de Zaragoza
000131433 245__ $$aFunneled Bayesian Optimization for Design, Tuning and Control of Autonomous Systems
000131433 260__ $$c2019
000131433 5203_ $$aIn this paper, we tackle several problems that appear in robotics and autonomous systems: algorithm tuning, automatic control, and intelligent design. All those problems share in common that they can be mapped to global optimization problems where evaluations are expensive. Bayesian optimization (BO) has become a fundamental global optimization algorithm in many problems where sample efficiency is of paramount importance. BO uses a probabilistic surrogate model to learn the response function and reduce the number of samples required. Gaussian processes (GPs) have become a standard surrogate model for their flexibility to represent a distribution over functions. In a black-box settings, the common assumption is that the underlying function can be modeled with a stationary GP. In this paper, we present a novel kernel function specially designed for BO, that allows nonstationary behavior of the surrogate model in an adaptive local region. This kernel is able to reconstruct nonstationarity even with the irregular sampling distribution that arises from BO. Furthermore, in our experiments, we found that this new kernel results in an improved local search (exploitation), without penalizing the global search (exploration) in many applications. We provide extensive results in well-known optimization benchmarks, machine learning hyperparameter tuning, reinforcement learning, and control problems, and UAV wing optimization. The results show that the new method is able to outperform the state of the art in BO both in stationary and nonstationary problems.
000131433 536__ $$9info:eu-repo/grantAgreement/ES/MINECO/DPI2015-65962-R
000131433 540__ $$9info:eu-repo/semantics/openAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/
000131433 590__ $$a11.079$$b2019
000131433 591__ $$aAUTOMATION & CONTROL SYSTEMS$$b1 / 63 = 0.016$$c2019$$dQ1$$eT1
000131433 591__ $$aCOMPUTER SCIENCE, CYBERNETICS$$b1 / 22 = 0.045$$c2019$$dQ1$$eT1
000131433 591__ $$aCOMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE$$b5 / 136 = 0.037$$c2019$$dQ1$$eT1
000131433 592__ $$a4.393$$b2019
000131433 593__ $$aElectrical and Electronic Engineering$$c2019$$dQ1
000131433 593__ $$aComputer Science Applications$$c2019$$dQ1
000131433 593__ $$aSoftware$$c2019$$dQ1
000131433 593__ $$aHuman-Computer Interaction$$c2019$$dQ1
000131433 593__ $$aInformation Systems$$c2019$$dQ1
000131433 593__ $$aControl and Systems Engineering$$c2019$$dQ1
000131433 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/publishedVersion
000131433 7102_ $$15007$$2520$$aUniversidad de Zaragoza$$bDpto. Informát.Ingenie.Sistms.$$cÁrea Ingen.Sistemas y Automát.
000131433 773__ $$g49, 4 (2019), 1489-1500$$pIEEE trans. cybern. (Print)$$tIEEE transactions on cybernetics$$x2168-2267
000131433 8564_ $$s1932136$$uhttps://zaguan.unizar.es/record/131433/files/texto_completo.pdf$$yVersión publicada
000131433 8564_ $$s3374304$$uhttps://zaguan.unizar.es/record/131433/files/texto_completo.jpg?subformat=icon$$xicon$$yVersión publicada
000131433 909CO $$ooai:zaguan.unizar.es:131433$$particulos$$pdriver
000131433 951__ $$a2024-02-09-14:28:11
000131433 980__ $$aARTICLE