Anderson Rocha, docente de Inteligência Artificial e Análises Forenses Digitais no Instituto de Computação da Universidade de Campinas (Unicamp), no Brasil, traz à Universidade da Beira Interior (UBI) “Reasoning for Complex Data through Self-Supervised Learning”, numa palestra sobre Inteligência Artificial e Machine Learning.
O docente, atual diretor do Instituto e do Laboratório de Inteligência Artificial (recod.ai) da Unicamp, partilhará o seu conhecimento com a comunidade ubiana no Departamento de Informática (Meeting Room), às 14h30, de 20 de julho (quinta-feira).
Abstract
Self-supervised learning deals with problems that have little or no available labeled data.
Recent work has shown impressive results when underlying classes have significant semantic differences. We will discuss strategies to tackle to enable learning from unlabeled data even when samples from different classes are not prominently diverse. We approach the problem by leveraging a novel ensemble-based clustering strategy where clusters derived from different configurations are combined to generate a better grouping for the data samples in a fully-unsupervised way. This strategy allows clusters with different densities and higher variability to emerge, which in turn reduces intraclass discrepancies, without requiring the burden of finding an optimal configuration per dataset. We will see results for Person Re-Identification and Text Authorship Verification but the techniques are useful in other applications as well.