Debate on the use of machine learning in computer networks closes 24th WRNP

- 23/05/2023

The last panel of the 24th WRNP debunked myths and pointed out promising paths for the use of machine learning (ML) in computer networks. Held jointly with the Brazilian Symposium on Computer Networks and Distributed Systems (SBRC), the table, mediated by Antonio Augusto de A. Rocha (UFF), was attended by researcher Damla Turgut, from the University of Central Florida (UCF) and by Bruno Ribeiro, from Purdue University. 

ML or machine learning is the ability of software to be trained and learn to adapt its own operation without human intervention. If it is true that machine learning is one of the main practical applications of artificial intelligence today, there are also expectations about what the technology is actually capable of doing in computer networks.

For Damla Turgut, two myths were created: that artificial intelligence only gives us perfect solutions and that AI is capable of replacing human work. “AI will actually help us become better at our jobs by providing the mechanisms we need to make smarter decisions based on data analysis”, said the researcher.

In the specific field of application for computer networks, Damla Turgut explained how machine learning is capable of optimizing the performance and efficiency of networks, detecting potential failures and improving the quality of service for the end user, both for common uses such as broadcasting football games and for more complex and individualized applications, such as smart homes.

Watch the entirety of the last WRNP panel: