Efficient Monte Carlo sampling for high-volume spaces and large medical and industrial databases

• Grant number: 14/23160-6
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date until: March 01, 2015
• End date until: April 30, 2016
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Francisco Louzada Neto
• Grantee: Luca Martino
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Associated research grant: 13/07375-0 - CeMEAI - Center for Mathematical Sciences Applied to Industry, AP.CEPID

Abstract

Monte Carlo (MC) methods are well-known computational techniques applied in different fields (as finance, biology, signal processing, machine learning) for Bayesian inference and stochastic optimization. Although they are very flexible methodologies and their applicability is almost unlimited, several drawbacks are still unsolved, specially for high dimensional problems and large data analysis. The main limitation is related to the computational cost and the time needed to obtain reliable results. In this project, we propose to investigate the combination of two main MC approaches, the importance sampling (IS) and the Markov Chain Monte Carlo (MCMC) schemes, considering jointly the use of parallel and population techniques. The developed methodology is intended to be applied to big industrial and medical data.

Scientific publications

(5)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

MARTINO, LUCA; ELVIRA, VICTOR; LOUZADA, FRANCISCO. Effective sample size for importance sampling based on discrepancy measures. Signal Processing, v. 131, p. 386-401, FEB 2017. (14/23160-6)

MARTINO, LUCA; READ, JESSE; ELVIRA, VICTOR; LOUZADA, FRANCISCO. Cooperative parallel particle filters for online model selection and applications to urban mobility. DIGITAL SIGNAL PROCESSING, v. 60, p. 172-185, JAN 2017. (14/23160-6)

MARTINO, L.; LOUZADA, F.. Issues in the Multiple Try Metropolis mixing. Computational Statistics, v. 32, n. 1, p. 239-252, MAR 2017. (14/23160-6)

MARTINO, L.; LOUZADA, F.. Adaptive rejection sampling with fixed number of nodes. COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, v. 48, n. 3, p. 655-665, MAR 16 2019. (14/23160-6)

MARTINO, L.; ELVIRA, V.; LUENGO, D.; CORANDER, J.; LOUZADA, F.. Orthogonal parallel MCMC methods for sampling and optimization. DIGITAL SIGNAL PROCESSING, v. 58, p. 64-84, NOV 2016. (14/23160-6)