Thesis Defense: Jesús López

Jesús López defended his PhD thesis work at the University of the Basque Country on October 18th 2018, being awarded Cum Laude mention (International Doctorate), and obtaining his degree in “Doctoral Programme in Mobile Network Information and Communication Technologies”.

Jesús’ thesis, which title was “New Perspectives and Methods for Stream Learning in the Presence of Concept Drift”, focused on Online Learning and Concept Drift issues, two of the hottest topics in the recent literature related to Machine Learning for Data Streams. They have revealed themselves as very relevant for the so-called Big Data paradigm, where nowadays we can find an increasing number of applications based on continuously produced training data, named as data streams. This thesis provides a complete overview of several related fields, tackling some open challenges that have been identified in the very recent state of the art. Concretely, it presents an innovative way to generate artificial diversity in ensembles, a set of necessary adaptations and improvements for evolving spiking neural networks in order to be used in online learning scenarios, and finally, a drift detector based on this former brain-inspired algorithm. All these approaches constitute an innovative work aimed at presenting new perspectives and methods in the field.

Supervisors: Javier Del Ser and Miren Nekane Bilbao

Evaluation Commitee: Albert Bifet (Télécom ParisTech), Sancho Salcedo-Sanz (University of Alcalá), and Manuel Vélez (University of the Basque Country)