Multiobjective optimization method for inverse problems and blind source separation

• Grant number: 14/27108-9
• Support Opportunities: Scholarships in Brazil - Master
• Start date: July 01, 2015
• End date: February 28, 2017
• Field of knowledge: Engineering - Electrical Engineering
• Agreement: Coordination of Improvement of Higher Education Personnel (CAPES)
• Principal Investigator: Leonardo Tomazeli Duarte
• Grantee: Guilherme Dean Pelegrina
• Host Institution: Faculdade de Ciências Aplicadas (FCA). Universidade Estadual de Campinas (UNICAMP). Limeira , SP, Brazil

Abstract

In signal and image processing there is a wide range of problems that can be formulated as an inverse problem or as a signal separation problem. Very often, these problems are solved by optimizing a criterion which, in the case of an inverse problem, can be a measure of distance between the observed data and the outputs of a considered model, and, in the case of signal separation, a separation criterion (contrast function). However, in many practical situations, there is some prior information that can be taken into account in order to obtain better results. Typically, the use of this prior information is carried out by simply adding terms to the adopted cost function, that is, via a mono-objective formulation. In this paper, though, we investigate an alternative approach in which the use of priori information is done by considering a multi-objective formulation. The main advantage is that, instead of an optimal solution, a multiobjective method provides a set of solutions (Pareto-optimal), which can be eventually submitted for consideration by the expert responsible for the process of decision making. Our study will focus on evolutionary multiobjective optimization methods and we shall consider experiments with synthetic data and real data acquired in problems related to chemical and seismic data. (AU)

Scientific publications

(The scientific publications listed on this page originate from the Web of Science or SciELO databases. Their authors have cited FAPESP grant or fellowship project numbers awarded to Principal Investigators or Fellowship Recipients, whether or not they are among the authors. This information is collected automatically and retrieved directly from those bibliometric databases.)

PELEGRINA, GUILHERME DEAN; ATTUX, ROMIS; DUARTE, LEONARDO TOMAZELI. Application of multi-objective optimization to blind source separation. EXPERT SYSTEMS WITH APPLICATIONS, v. 131, p. 60-70, OCT 1 2019. (14/27108-9, 15/16325-1)

PELEGRINA, GUILHERME D.; DUARTE, LEONARDO T.. A Multi-Objective Approach for Post-Nonlinear Source Separation and Its Application to Ion-Selective Electrodes. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, v. 65, n. 12, p. 2067-2071, DEC 2018. (15/16325-1, 14/27108-9)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

PELEGRINA, Guilherme Dean. A multi-objective optimization approach for blind source separation. 2017. Master's Dissertation - Universidade Estadual de Campinas (UNICAMP). Faculdade de Ciências Aplicadas Limeira, SP.