Use of Meta-learning to improve deep learning algorithms in classification problems

• Grant number: 15/03986-0
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: October 01, 2015
• End date: September 30, 2016
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: André Carlos Ponce de Leon Ferreira de Carvalho
• Grantee: Rafael Gomes Mantovani
• Supervisor: Joaquin Vanschoren
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: Eindhoven University of Technology (TU/e), Netherlands
• Associated to the scholarship: 12/23114-9 - Use of meta-learning for parameter tuning for classification problems, BP.DR

Abstract

Machine learning approaches have been widely effective solving many simple and well de- fined problems in the literature. However, most approaches face difficulties when dealing with more complicated real-world problems, often due to high data dimensionality. One of the first applications of Deep Learning models was with data dimensionality reduction. These learning models have attracted a substantial amount of academic and industrial attention in the past years, and are the current state-of-the-art in many applications such as speech recognition, image pedestrian detection, image classification as well as many medical applications. Although building and training deep learning models is highly de- sirable, there lies a big difficulty in selecting the correct parameter settings that needs to be better understood. In this project, we aim to investigate how to combine meta-learning with optimization techniques to efficiently build Deep Learning models for classification tasks. (AU)

Scientific publications

(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)

MANTOVANI, RAFAEL G.; ROSSI, ANDRE L. D.; ALCOBACA, EDESIO; VANSCHOREN, JOAQUIN; DE CARVALHO, ANDRE C. P. L. F.. A meta-learning recommender system for hyperparameter tuning: Predicting when tuning improves SVM classifiers. INFORMATION SCIENCES, v. 501, p. 193-221, OCT 2019. (12/23114-9, 15/03986-0, 18/14819-5)

MANTOVANI, RAFAEL G.; HORVATH, TOMAS; CERRI, RICARDO; VANSCHOREN, JOAQUIN; DE CARVALHO, ANDRE C. P. L. F.; IEEE. Hyper-parameter Tuning of a Decision Tree Induction Algorithm. PROCEEDINGS OF 2016 5TH BRAZILIAN CONFERENCE ON INTELLIGENT SYSTEMS (BRACIS 2016), v. N/A, p. 6-pg., 2016-01-01. (13/07375-0, 15/03986-0, 12/23114-9)