Matheus H. N. Nunes

Research Assistant

Researching about machine learning methods for graph data.

• Nunes, Matheus and Pappa, Gisele L., Neural Architecture Search in Graph Neural Networks , on 9th Brazilian Conference on Intelligent Systems (BRACIS) - 2020

Software Engineering Intern

C++ data mining for the Health Search team. Details under NDA.

Research and Development (R & D) Intern

Collected, analyzed and produced visualizations of approximately 10 terabytes of Wi-Fi traffic data using R and MATLAB. Built a model for predicting for Quality of Service (QoS) metrics (throughput, delay, jitter and packet loss) on Wi-Fi stations from Wi-Fi APs’ data, using Recurrent Neural Networks (GRU) in Python, achieving a RMSE of the order of 10−2 for throughput, 10−4 for delay, and 10−5 for jitter and packet loss. Work resulted in paper published at PIMRC 2019.

Developed and deployed a Cloud Computing system as an IaaS platform for the UFMG Testbed within months, which led to UFMG being the first Brazilian university in project FUTEBOL federated by FED4FIRE. Built solutions based on Docker and LXD containers, for virtualizing networking devices. Developed a voice recognition application for testing and demonstrating the use of Testbed resources. Tool resulted in paper published at INFOCOM 2017. Planned and implemented the network infrastructure of the lab for the project. Configured, installed and maintained Linux servers and was responsible for controlling the software versioning of the project.

• Moura, Henrique, Nunes, Matheus, Miranda Jr, Gilson., Correia, Luiz H. A., Macedo, Daniel F., Estimating Quality of Service on Wi-Fi Stations Using Recurrent Neural Networks , on the 30th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) - 2019
• Silva, Aloizio P. et. al., Impact of Fog and Cloud Computing on an IoT Service Running Over an Optical/Wireless Network Testbed , on 2017 IEEE Conference on Computer Communications Workshops (INFOCOM Workshops, CNERT)
• Silva, Aloizio P. et. al., Demo abstract: Assessing the Impact of Fog and Cloud Computing on an IoT Service Running Over an Optical/Wireless Network - An experimental approach , on 2017 IEEE Conference on Computer Communications Workshops (INFOCOM Workshops)

Research Assistant

Developed TSViz, a standalone web app for visualizing multi-threaded systems' execution logs, built using D3.js. Work resulted in paper published at ICSE 2017.

• Nunes, Matheus. et. al., Studying Multi-threaded Behavior with TSViz , on 2017 IEEE/ACM 39th International Conference on Software Engineering Companion (ICSE-C)

Universidade Federal de Minas Gerais (UFMG)

GPA: 9.3/10

• Head of students, on the Graduate Council

Universidade Federal de Minas Gerais (UFMG)

GPA: 84/100

Take a look at some of my class projects!