2,742 Hits in 10.4 sec

Finite Sample Convergence Rates of Zero-Order Stochastic Optimization Methods

John C. Duchi, Michael I. Jordan, Martin J. Wainwright, Andre Wibisono
2012 Neural Information Processing Systems  
We show that if pairs of function values are available, algorithms that use gradient estimates based on random perturbations suffer a factor of at most √ d in convergence rate over traditional stochastic  ...  We consider derivative-free algorithms for stochastic optimization problems that use only noisy function values rather than gradients, analyzing their finite-sample convergence rates.  ...  JCD was also supported by an NDSEG fellowship and a Facebook PhD fellowship.  ... 
dblp:conf/nips/DuchiJWW12 fatcat:cwbm4en7czenbimttld4qarkz4

Feature Article: Optimization for simulation: Theory vs. Practice

Michael C. Fu
2002 INFORMS journal on computing  
The main thesis of this article, however, is that there is a disconnect between research in simulation optimization-which has addressed the stochastic nature of discrete-event simulation by concentrating  ...  P robably one of the most successful interfaces between operations research and computer science has been the development of discrete-event simulation software.  ...  Acknowledgments This work was supported in part by the National Science Foundation under Grants DMI-9713720 and DMI-9988867 and by the Air Force Office of Scientific Research under Grant F496200110161.  ... 
doi:10.1287/ijoc. fatcat:ntus2mrdhjbptc77jf2zjmgmcu

A fuzzy simulated evolution algorithm for integrated manufacturing system design

Michael Mutingi
2013 International Journal of Industrial Engineering Computations  
This study seeks to develop a fuzzy simulated evolution algorithm (FSEA) that integrates fuzzy-set theoretic concepts and the philosophy of constructive perturbation and evolution.  ...  Illustrative computational experiments based on existing problem instances from the literature demonstrate the utility and the strength of the FSEA algorithm developed in this study.  ...  Computational results based on a set of benchmark problems concerned with the cell formation problem from the literature revealed that FSEA is a competitive algorithm which can solve hard combinatorial  ... 
doi:10.5267/j.ijiec.2013.01.003 fatcat:g4ebtzajtfb4zcty2a6y7qnhhe

The self-learning AI controller for adaptive power beaming with fiber-array laser transmitter system [article]

A.M. Vorontsov, G.A. Filimonov
2022 arXiv   pre-print
For optimization of power transition through the atmosphere fiber-array is traditionally controlled by stochastic parallel gradient descent (SPGD) algorithm where control feedback is provided via radio  ...  A DNN training is occurred online in sync with control system operation and is performed by applying of small perturbations to DNN's outputs.  ...  Vorontsov for useful discussions of fiber-array power beaming problem and A.B. Lavrentyev for the interest in AI-based control systems.  ... 
arXiv:2204.05227v1 fatcat:ubmfmgztqjahzj5grwnk7fjdrm

The Real Deal: A Review of Challenges and Opportunities in Moving Reinforcement Learning-Based Traffic Signal Control Systems Towards Reality [article]

Rex Chen, Fei Fang, Norman Sadeh
2022 arXiv   pre-print
However, more work should take a systems thinking approach that considers the impacts of other pipeline components on RL.  ...  We focus on four challenges involving (1) uncertainty in detection, (2) reliability of communications, (3) compliance and interpretability, and (4) heterogeneous road users.  ...  of this review.  ... 
arXiv:2206.11996v2 fatcat:jwwnydngozegjcuvsbfzodqpfi

Development of an Interactive AI System for the Optimal Timing Prediction of Successful Weaning from Mechanical Ventilation for Patients in Respiratory Care Centers

Kuang-Ming Liao, Shian-Chin Ko, Chung-Feng Liu, Kuo-Chen Cheng, Chin-Ming Chen, Mei-I Sung, Shu-Chen Hsing, Chia-Jung Chen
2022 Diagnostics  
This study aims to utilize artificial intelligence algorithms to build predictive models for the successful timing of the weaning of patients from MV in RCCs and to implement a dashboard with the best  ...  The development of an AI prediction dashboard is a promising method to assist in the prediction of the optimal timing of weaning from MV in RCC settings.  ...  The AI Center and Department of Information Systems of Chi Mei embedded the XGBoost model in a web-based system (digital dashboard) for predicting the optimal MV weaning timing for patients in the RCC  ... 
doi:10.3390/diagnostics12040975 pmid:35454023 pmcid:PMC9030191 fatcat:f6vsr6pdq5gezdmdiaxz6fpcl4

Retrospective-approximation algorithms for the multidimensional stochastic root-finding problem

Raghu Pasupathy, Bruce W. Schmeiser
2009 ACM Transactions on Modeling and Computer Simulation  
This paper presents a family of algorithms to solve the multidimensional (q ≥ 1) SRFP, generalizing Chen and Schmeiser's onedimensional retrospective approximation (RA) family of algorithms.  ...  We focus on a specific member of the family, called the Bounding RA algorithm, which finds a sequence of polytopes that progressively decrease in size while approaching the solution.  ...  See Chen (1994) for more on (1), (2), (3) and (4). COMPLICATING ISSUES The SRFP is one stochastic generalization of the problem of solving a non-linear deterministic system of equations.  ... 
doi:10.1145/1502787.1502788 fatcat:iogtlriohnhl7p7mvwlsxgf42i

Autonomous discovery in the chemical sciences part I: Progress

Klavs F. Jensen, Connor W Coley, Natalie S Eyke
2019 Angewandte Chemie International Edition  
Part two reflects on these case studies and identifies a set of open challenges for the field.  ...  We then introduce a set of questions and considerations relevant to assessing the extent of autonomy.  ...  Acknowledgements We thank Thomas Struble for providing comments on the manuscript and our other colleagues and collaborators for useful conversations around this topic.  ... 
doi:10.1002/anie.201909987 pmid:31553511 fatcat:yfg4jnixhvfgdmnr4p6mbyolpe

A simulation-based approach to capturing autocorrelated demand parameter uncertainty in inventory management

Alp Akcay, Bahar Biller, Sridhar Tayur
2012 Proceedings Title: Proceedings of the 2012 Winter Simulation Conference (WSC)  
Using a simulation-based sampling algorithm, we quantify the expected cost due to parameter uncertainty as a function of the length of the historical demand data, the critical fractile, the parameters  ...  We consider a repeated newsvendor setting where the parameters of the demand distribution are unknown, and we study the problem of setting inventory targets using only a limited amount of historical demand  ...  We approximate the expected cost due to demand parameter uncertainty using a simulation-based sampling algorithm.  ... 
doi:10.1109/wsc.2012.6465035 dblp:conf/wsc/AkcayBT12 fatcat:qeskm75ag5dadiuqse2pcfiru4

Simulation optimization: A tutorial overview and recent developments in gradient-based methods

Marie Chau, Michael C. Fu, Huashuai Qu, Ilya O. Ryzhov
2014 Proceedings of the Winter Simulation Conference 2014  
2.2 is based heavily on material in Chau and Fu (2014).  ...  We then describe some recent research in two areas of simulation optimization: stochastic approximation and the use of direct stochastic gradients for simulation metamodels.  ...  Specifically Chapter 2 presents some locally and globally convergent algorithms for discrete optimization on a large countable (possibly infinite) set, and Chapter 12 presents iterative model-based algorithms  ... 
doi:10.1109/wsc.2014.7019875 dblp:conf/wsc/ChauFQR14 fatcat:jzkyzx6kbnfephf6xtmea7dxoq

Design of large polyphase filters in the Quadratic Residue Number System

Gian Carlo Cardarilli, Alberto Nannarelli, Yann Oster, Massimo Petricca, Marco Re
2010 2010 Conference Record of the Forty Fourth Asilomar Conference on Signals, Systems and Computers  
We conjecture, based on our simulation results, that the LRT rule is always optimal both in terms of robustness and error probability.  ...  This paper studies the optimization of a multicell MIMO downlink system in which each base station serves multiple users, and each user is served by only one base station.  ... 
doi:10.1109/acssc.2010.5757589 fatcat:ccxnu5owr5fyrcjcqukumerueq

Optimal disease outbreak decisions using stochastic simulation

Mike Ludkovski, Jarad Niemi
2011 Proceedings of the 2011 Winter Simulation Conference (WSC)  
We present a methodology for dynamic determination of optimal policies in a stochastic compartmental model with parameter uncertainty.  ...  The latter step is simulation-based and relies on regression Monte Carlo techniques. To improve performance we investigate lasso regression and global policy iteration.  ...  ACKNOWLEDGEMENTS Ludkovski's research was partially supported by the Hellman Family Foundation Grant.  ... 
doi:10.1109/wsc.2011.6148076 dblp:conf/wsc/LudkovskiN11 fatcat:du53tr64krhfbcffkvny2kd4ie

Unconstraining methods for revenue management systems under small demand

Nikolaos Kourentzes, Dong Li, Arne K. Strauss
2017 Journal of Revenue and Pricing Management  
Empirical results on accuracy and revenue performance based on data from a major car rental company indicate revenue improvements over a best practice benchmark by statistically significant 0.5%-1.4% in  ...  Furthermore, they are numerically robust due to our proposed group-based parameter optimization.  ...  Acknowledgement All authors contributed equally and are named in alphabetical order.  ... 
doi:10.1057/s41272-017-0117-x fatcat:qb7hp4td4rf6pgqujgwbwohz74

A Simulation Optimization Approach to Epidemic Forecasting

Elaine O. Nsoesie, Richard J. Beckman, Sara Shashaani, Kalyani S. Nagaraj, Madhav V. Marathe, Man-Seong Park
2013 PLoS ONE  
This study represents the final step of a project aimed at using a combination of simulation, classification, statistical and optimization techniques to forecast the epidemic curve and infer underlying  ...  Reliable forecasts of influenza can aid in the control of both seasonal and pandemic outbreaks. We introduce a simulation optimization (SIMOP) approach for forecasting the influenza epidemic curve.  ...  Acknowledgments We thank our external collaborators and members of the Network Dynamics and Simulation Science Laboratory (NDSSL) for their suggestions and comments.  ... 
doi:10.1371/journal.pone.0067164 pmid:23826222 pmcid:PMC3694918 fatcat:d2ouvfwtlfelljfkijl6afr5pa

Integer-Ordered Simulation Optimization using R-SPLINE

Honggang Wang, Raghu Pasupathy, Bruce W. Schmeiser
2013 ACM Transactions on Modeling and Computer Simulation  
We consider simulation-optimization (SO) models where the decision variables are integer ordered and the objective function is defined implicitly via a simulation oracle, which for any feasible solution  ...  We develop R-SPLINE-a Retrospective-search algorithm that alternates between a continuous Search using Piecewise-Linear Interpolation and a discrete Neighborhood Enumeration, to asymptotically identify  ...  We thank the area editor, associate editor and the two referees for the high quality of reviews.  ... 
doi:10.1145/2499913.2499916 fatcat:liczv6ulsbhhnm6ubbmgaarptu
« Previous Showing results 1 — 15 out of 2,742 results