Learn More
The advantages of discriminative learning algorithms and kernel machines are combined with generative modeling using a novel kernel between distributions. In the probability product kernel, data points in the input space are mapped to distributions over the sample space and a general inner product is then evaluated as the integral of the product of pairs of(More)
We propose a new technique for direct visual matching of images for the purposes of face recognition and image retrieval, using a probabilistic measure of similarity, based primarily on a Bayesian (MAP) analysis of image di!erences. The performance advantage of this probabilistic matching technique over standard Euclidean nearest-neighbor eigenface matching(More)
Tony Jebara MIT Media Lab 20 Ames St. Cambridge, MA 02139 jebara@media. mit. edu We present a general framework for discriminative estimation based on the maximum entropy principle and its extensions. All calculations involve distributions over structures and/or parameters rather than specific settings and reduce to relative entropy projections. This holds(More)
In various application domains, including image recognition, it is natural to represent each example as a set of vectors. With a base kernel we can implicitly map these vectors to a Hilbert space and fit a Gaussian distribution to the whole set using Kernel PCA. We define our kernel between examples as Bhattacharyya’s measure of affinity between such(More)
centered on a large database, but in this case it is entirely of living organisms, the marine bivalves. Over 28,000 records of bivalve genera and subgenera from 322 locations around the world have now been compiled by these authors, giving a global record of some 854 genera and subgenera and 5132 species. No fossils are included in the database, but because(More)
Graph based semi-supervised learning (SSL) methods play an increasingly important role in practical machine learning systems. A crucial step in graph based SSL methods is the conversion of data into a weighted graph. However, most of the SSL literature focuses on developing label inference algorithms without extensively studying the graph building method(More)
Graph transduction methods label input data by learning a classification function that is regularized to exhibit smoothness along a graph over labeled and unlabeled samples. In practice, these algorithms are sensitive to the initial set of labels provided by the user. For instance, classification accuracy drops if the training set contains weak labels, if(More)
In their day-to-day lives, people naturally understand and operate in a three dimensional world. Curiously, though, they only sense 2D projections of it. The seemingly e ortless act of inferring 3D from 2D observations is the result of complex mechanisms that are still quite far from being resolved. For many years, this task has been considered the primary(More)