Giovani in un'ora - Ciclo di seminari - Terza parte
- Day - Time: 18 March 2021, h.11:00
- Place: Area della Ricerca CNR di Pisa - Room: Zoom
Speakers
Referent
Abstract
Riccardo Guidotti - Explaining Any Time Series Classifier
Abstract: We present LASTS, a method to explain the decisions of black box models for time series classification. The explanation consists of factual and counterfactual shapelet-based rules revealing the reasons for the classification, and of a set of exemplars and counter-exemplars highlighting similarities and differences with the time series under analysis. LASTS first generates exemplar and counter-exemplar time series in the latent feature space and learns a local latent decision tree classifier. Then, it selects and decodes those respecting the decision rules explaining the decision. Finally, it learns on them a shapelet-tree that reveals the parts of the time series that must, and must not, be contained for getting the returned outcome from the black box. A wide experimentation shows that LASTS provides faithful, meaningful and interpretable explanations.
Felice Manlio Bacco - Multipath techniques to improve the quality of experience of real-time multimedia traffic in mobility conditions
Abstract: In this seminar, the use of multipath techniques will be presented as a means to improve the quality of experience of multimedia traffic in the case of UAV flights in beyond-visual-line-of-sight (BVLoS) conditions. In the real testbed that will be presented, multiple cellular connections have been jointly used to deliver a real-time video feed from an UAV to a ground station to assist pilot and flight operations. The aim is in providing reliable contextual awareness by exploiting techniques providing both redundancy and reliability aiming at enhancing the user experience.