CV

ANT
ˆ
ONIO H. RIBEIRO
Belo Horizonte, MG Brazil
antonio-ribeiro@ufmg.br antonior92.github.io
Ph.D. candidate in Electrical Engineering at UFMG. Interested in system identification, machine learning
and optimization.
EDUCATION
Ph.D., Electrical Engineering Aug. 2017 -
Universidade Federal de Minas Gerais (UFMG)
Supervisor: Luis Antonio Aguirre
M.S., Electrical Engineering Jan. 2016 - Jul. 2017
Universidade Federal de Minas Gerais (UFMG)
Title: Recurrent Structures in System Identification
Supervisor: Luis Antonio Aguirre
GPA: 5.00/5.00
B.S.E., Electrical Engineering Jan. 2011 - Jan. 2016
Universidade Federal de Minas Gerais (UFMG)
GPA: 4.91/5.00
WORK EXPERIENCE
Google Summer of Code May. 2017 - Aug. 2017
Software Developer Scipy
· I have successfully completed Google Summer of Code program under the mentorship of Matt Haber-
land, Nikolay Mayorov and Ralf Gommers. My project was the implementation of an interior-point
solver for large-scale nonlinear programming problems. The implementation is available in the GitHub
repository https://github.com/antonior92/ip-nonlinear-solver and is being integrated to SciPy,
the Python scientific library, through the pull request #7729.
Invent Vision Jan. 2015 - Dec. 2015
Hardware Team Intern Belo Horizonte, Brazil
· I was part of the hardware development team and worked designing FPGA-based cameras. The major
project I have worked on while there was the design and implementation of a stereo camera.
R&D Project with Petrobras Oil Company Jun. 2013 - Jan. 2015
Undergraduate Researcher Belo Horizonte, MG
· I worked on the development of methods for identification of oil well mathematical models under the
supervision of Luis Antonio Aguirre.
COMPUTER SKILLS
Languages C/C++, Python, Matlab, Julia and R.
PUBLICATIONS
2017
· Lasso Regularization Paths for NARMAX Models via Coordinate Descent - Preprint (arXiv:1710.00598).
Anonio H. Ribeiro, Luis A. Aguirre. Submitted to American Control Conference.
· “Parallel Training Considered Harmful?”: Comparing Series-Parallel and Parallel Feedforward Net-
work Training - Preprint (arXiv:1706.07119). Anonio H. Ribeiro, Luis A. Aguirre. Submitted to
Neurocomputing.
· Shooting Methods for Parameter Estimation of Output Error Models - Proceedings of IFAC World
Congress (Toulouse, France - 2017). Anonio H. Ribeiro, Luis A. Aguirre.
2015
· Selecting Transients Automatically for the Identification of Models for an Oil Well - Proceedings of
2nd IFAC Workshop on Automatic Control in Offshore Oil and Gas Production (Florianpolis, Brazil -
2015). Antˆonio H. Ribeiro, Luis A. Aguirre.
2014
· Rela¸oes Est´aticas de modelos NARX MISO e sua Representa¸ao de Hammerstein - Proceedings of XX
Brazilian Congress of Automatica (Belo Horizonte, Brazil - 2014). Antˆonio H. Ribeiro, Luis A. Aguirre.
OPEN-SOURCE CONTRIBUTIONS
Contributions to Scipy
· I am a member of SciPy core development team, having contributed to optimization and signal pro-
cessing packages with the implementation of IIR filters (implementation of functions iirnotch and
irrpeak), trust-region optimization methods (implementation of method trust-exact) and other gen-
eral improvements (such as modifications in the BFGS implementation to improve its performance).
Check my GitHub account: https://github.com/antonior92 to see a complete list of my open-source
contributions.