Rodrigo Mello is Associate Professor at the Institute of Mathematical and Computer Sciences, Department of Computer Science, University of São Paulo, São Carlos, SP, Brazil. He obtained his doctoral degree in 2003 from the University of São Paulo. His research interests include machine learning, applications in dynamic systems, time series analysis and data streams.
Have you ever come across the need of proving that your supervised learning algorithm indeed learned something from data? How could you define the minimal training sample size to ensure learning guarantees to your algorithm on some target problem? This will be a provocative talk on such a topic with the aim of bringing some initial thoughts/conclusions on it.
This event is co-organized with the Machine Learning Group at University of Waikato.