1,721,406 research outputs found
Regularization, Adaptation and Generalization of Neural Networks
No full textThe ability to generalize to unseen data is one of the fundamental, desired
properties in a learning system. This thesis reports dierent research eorts
in improving the generalization properties of machine learning systems at
dierent levels, focusing on neural networks for computer vision tasks.
First, a novel regularization method is presented, Curriculum Dropout. It
combines Curriculum Learning and Dropout, and shows better regularization
eects than the original algorithm in a variety of tasks, without requiring
substantially any additional implementation eorts.
While regularization methods are extremely powerful to better generalize
to unseen data from the same distribution as the training one, they are not
very successful in mitigating the dataset bias issue. This problem constitutes
in models learning the peculiarities of the training set, and poorly generalizing
to unseen domains. Unsupervised domain adaptation has been one of the main
solutions to this problem. Two novel adaptation approaches are presented in
this thesis. First, we introduce the DIFA algorithm, which combines domain
invariance and feature augmentation to better adapt models to new domains
by relying on adversarial training. Next, we propose an original procedure that
exploits the \mode collapse" behavior of Generative Adversarial Networks.
Finally, the general applicability of domain adaptation algorithms is
questioned (due to the assumptions of knowing the target distribution a
priori and being able to sample from it). A novel framework is presented to
overcome its liabilities, where the goal is to generalize to unseen domains by
relying only on data from a single source distribution. We face this problem
through the lens of robust statistics, dening a worst-case formulation where
the model parameters are optimized with respect to populations which are
-distant from the source domain on a semantic space
Addressing Model Vulnerability to Distributional Shifts Over Image Transformation Sets
No full textWe are concerned with the vulnerability of computer vision models to distributional shifts. We formulate a combinatorial optimization problem that allows evaluating the regions in the image space where a given model is more vulnerable, in terms of image transformations applied to the input, and face it with standard search algorithms. We further embed this idea in a training procedure, where we define new data augmentation rules according to the image transformations that the current model is most vulnerable to, over iterations. An empirical evaluation on classification and semantic segmentation problems suggests that the devised algorithm allows to train models that are more robust against content-preserving image manipulations and, in general, against distributional shifts(1)
Considerazioni sperimentali e clinico terapeutiche su un preparato cronopluricorticoide alla luce della cronobiologia.
No full textAbnormal growth hormone and cortisol but not thyroid stimulating hormone response to an intravenous glucose tolerance test in normal weight bulimic women
No full textEffetto coleretico e colagogo dell'acqua sulfureo-solfato calcica di Tabiano sulla colestasi nell'epatopatia alcool-correlata
No full text
- …
