DEI/CISUC Seminars

Publication Date: 2018-11-20 15:08:43

November 21, Wednesday,
13h (sharp),
Room A.5.4. DEI-FCTUC

Invited Speaker: João Correia

Title: EFECTIVE - Evolutionary Framework for Classifier Assessment and Improvement


Typical Machine Learning approaches rely on a dataset and a model to solve problems. For most problems, optimisation of Machine Learning approaches is crucial to attaining competitive performances. Most of the effort goes toward optimising the model by exploring new algorithms and tuning the parameters. Nevertheless, the dataset is also a key part for Machine Learning performance. Gathering, constructing and optimising a representative dataset is a hard task and a time-consuming endeavour, with no well-established guidelines to follow. The Evolutionary Framework for Classifier Assessment and Improvement is introduced and explored. The key parts of the framework are the following: the Classifier System module, which holds the Machine Learning model that is going to be assessed and improved; the Evolutionary Computation module responsible for generating the new instances using the Classifier System module for fitness.

Short bio: João Correia is an Invited Assistant at the University of Coimbra and a researcher of the Computational Design and Visualization Lab. from the Cognitive Media Systems group of the Centre for Informatics and Systems of the same university. He holds a PhD in Information Science and Technology from the University of Coimbra. He also holds a MSc and BS in Informatics Engineering from the University of Coimbra. His main research interests include Evolutionary Computation, Machine Learning, Pattern Recognition, Computer Vision and Computational Creativity.