Interpretability and Multi-Objective HS Methods for Feature Selection

In this project, we are working on developing new approaches that allow the extraction of implicit and hidden knowledge in the huge mass of existing data in different application domains. The domains studied cover, but are not limited to, problems in Bioinformatics, allowing us to understand better the functional and structural aspects of how genes and proteins and their relationships with pathological phenotypes. Machine learning techniques can efficiently deal with non-linear and noisy environments and adequately handle missing data. However, machine learning models cannot always be easily interpreted. Interpretable machine learning is the collection of algorithms and techniques that allow humans to understand the cause of a decision made by a machine learning model. This research project aims to advance the area of Machine Learning and the Application of these techniques to enable knowledge discovery in relevant scientific problems.

Dr. Márcio Dorn - Coordinator
PPGC/INF/UFRGS - Brazil

Dr. Mario Inostroza-Ponta - Coordinator
DIINF/USACH - Chile

Dr. Bruno César Feltes
PPGC/INF/UFRGS - Brazil

Dra. Joice de Faria Poloni
PPGC/INF/UFRGS - Brazil

Dr. Manuel Escalona
PPGC/INF/UFRGS - Brazil

Dr. Rodrigo Ligabue Braun
DF/UFCSPA - Brazil

Mateus Boiani
Ph.D. Candidate
PPGC/INF/UFRGS - Brazil

Bruno Iochins Grisci
Ph.D. Candidate
PPGC/INF/UFRGS - Brazil

Gabriel Dominico
Ph.D. Candidate
PPGC/INF/UFRGS - Brazil

Éderson S. M. Pinto
Ph.D. Candidate
PPGBCM/CBIOT/UFRGS - Brazil

Flavielle B. Marques
Ph.D. Candidate
PPGC/INF/UFRGS - Brazil

Jonas da S. Bohrer
Ph.D. Candidate
PPGC/INF/UFRGS - Brazil

Leonardo da Luz Dorneles
M.Sc. Candidate
PPGC/INF/UFRGS - Brazil

Gabriela F. Gonçalves
Ph.D. Candidate
PPGBCM/CBIOT/UFRGS - Brazil

Régis Antonioli Jr
M.Sc. Candidate
PPGBCM/CBIOT/UFRGS - Brazil