F
Fabio Gagliardi Cozman
Researcher at University of São Paulo
Publications - 248
Citations - 4447
Fabio Gagliardi Cozman is an academic researcher from University of São Paulo. The author has contributed to research in topics: Probabilistic logic & Bayesian network. The author has an hindex of 30, co-authored 230 publications receiving 4191 citations. Previous affiliations of Fabio Gagliardi Cozman include Carnegie Mellon University.
Papers
More filters
Journal ArticleDOI
Credal networks
TL;DR: Credal networks as mentioned in this paper is a compact representation for a set of probability distributions, and it is closely related to very popular statistical models such as Markov chains, Bayesian networks, Markov random fields, etc.
Journal ArticleDOI
Semisupervised learning of classifiers: theory, algorithms, and their application to human-computer interaction
TL;DR: A new analysis is provided that shows under what conditions unlabeled data can be used in learning to improve classification performance, and how the resulting algorithms are successfully employed in two applications related to human-computer interaction and pattern recognition: facial expression recognition and face detection.
Proceedings Article
Semi-supervised learning of mixture models
TL;DR: This paper analyzes the performance of semi-supervised learning of mixture models and shows that unlabeled data can lead to an increase in classification error even in situations where additional labeled data would decrease classification error.
Proceedings ArticleDOI
Depth from scattering
TL;DR: This work investigates a group of techniques for extraction of depth cues solely from the analysis of atmospheric scattering effects in images, and finds that depth cues in outdoor scenes can be recovered with surprising accuracy and can be used as an additional information source for autonomous vehicles.
Proceedings Article
Unlabeled Data Can Degrade Classification Performance of Generative Classifiers
Fabio Gagliardi Cozman,Ira Cohen +1 more
TL;DR: It is shown that unlabeled data can degrade the performance of a classifier when there are discrepancies between modeling assumptions used to build the classifier and the actual model that generates the data.