Zobrazeno 1 - 10
of 16
pro vyhledávání: '"Polito, Marzia"'
Autor:
Dukler, Yonatan, Achille, Alessandro, Paolini, Giovanni, Ravichandran, Avinash, Polito, Marzia, Soatto, Stefano
We present a method to compute the derivative of a learning task with respect to a dataset. A learning task is a function from a training set to the validation error, which can be represented by a trained deep neural network (DNN). The "dataset deriv
Externí odkaz:
http://arxiv.org/abs/2111.09785
Autor:
Li, Zhizhong, Ravichandran, Avinash, Fowlkes, Charless, Polito, Marzia, Bhotika, Rahul, Soatto, Stefano
Traditionally, distillation has been used to train a student model to emulate the input/output functionality of a teacher. A more useful goal than emulation, yet under-explored, is for the student to learn feature representations that transfer well t
Externí odkaz:
http://arxiv.org/abs/2107.08039
Autor:
Majumder, Orchid, Ravichandran, Avinash, Maji, Subhransu, Achille, Alessandro, Polito, Marzia, Soatto, Stefano
Few-shot learning aims to transfer information from one task to enable generalization on novel tasks given a few examples. This information is present both in the domain and the class labels. In this work we investigate the complementary roles of the
Externí odkaz:
http://arxiv.org/abs/2101.11058
Autor:
Golatkar, Aditya, Achille, Alessandro, Ravichandran, Avinash, Polito, Marzia, Soatto, Stefano
We show that the influence of a subset of the training samples can be removed -- or "forgotten" -- from the weights of a network trained on large-scale image classification tasks, and we provide strong computable bounds on the amount of remaining inf
Externí odkaz:
http://arxiv.org/abs/2012.13431
Autor:
Achille, Alessandro, Golatkar, Aditya, Ravichandran, Avinash, Polito, Marzia, Soatto, Stefano
Classifiers that are linear in their parameters, and trained by optimizing a convex loss function, have predictable behavior with respect to changes in the training data, initial conditions, and optimization. Such desirable properties are absent in d
Externí odkaz:
http://arxiv.org/abs/2012.11140
Autor:
Polito, Marzia
We give a complete description of the tautological subgroup of the fourth cohomology group of the moduli space of pointed stable curves, and prove that for g \geq 8 it coincides with the cohomology group itself. We further give a conjectural upper bo
Externí odkaz:
http://arxiv.org/abs/math/0005129
Autor:
Gleich, David, Polito, Marzia
Publikováno v:
Internet Math. 3, no. 3 (2006), 257-294
In this paper, we consider the problem of calculating fast and accurate approximations to the personalized PageRank score of a webpage. We focus on techniques to improve speed by limiting the amount of web graph data we need to access. ¶ Our algorit
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=project_eucl::8d6fe4d761fc2b95be13d61b71104e01
http://projecteuclid.org/euclid.im/1204906158
http://projecteuclid.org/euclid.im/1204906158
Consider a number of moving points, where each point is attached to a joint of the human body and projected onto an image plane. Johannson showed that humans can effortlessly detect and recognize the presence of other humans from such displays. This
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od________38::78a3bebef9a24c61c79a570b3156909e
https://resolver.caltech.edu/CaltechAUTHORS:20140730-101718695
https://resolver.caltech.edu/CaltechAUTHORS:20140730-101718695
Autor:
Polito, Marzia, Perona, Pietro
Locally Linear Embedding (LLE) is an elegant nonlinear dimensionality-reduction technique recently introduced by Roweis and Saul 2]. It fails when the data is divided into separate groups. We study a variant of LLE that can simultaneously group the d
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od________38::1b26cf5499c2d30b175813092b45a4bf
https://resolver.caltech.edu/CaltechAUTHORS:20140730-101719764
https://resolver.caltech.edu/CaltechAUTHORS:20140730-101719764
Autor:
Yi, Haoran, Kozintsev, Igor, Polito, Marzia, Wu, Yi, Bouguet, Jean-Yves, Nefian, Ara, Dulong, Carole
Publikováno v:
Proceedings of SPIE; Nov2007, Issue 1, p65060Q-65060Q-12, 12p
