Zobrazeno 1 - 7
of 7
pro vyhledávání: '"Pouplin, Alison"'
One of the main challenges in modern deep learning is to understand why such over-parameterized models perform so well when trained on finite data. A way to analyze this generalization concept is through the properties of the associated loss landscap
Externí odkaz:
http://arxiv.org/abs/2307.04719
Riemannian geometry provides us with powerful tools to explore the latent space of generative models while preserving the underlying structure of the data. The latent space can be equipped it with a Riemannian metric, pulled back from the data manifo
Externí odkaz:
http://arxiv.org/abs/2212.10010
Autor:
Scherer, Paul, Gaudelet, Thomas, Pouplin, Alison, Del Vecchio, Alice, S, Suraj M, Bolton, Oliver, Soman, Jyothish, Taylor-King, Jake P., Edwards, Lindsay
In constrained real-world scenarios, where it may be challenging or costly to generate data, disciplined methods for acquiring informative new data points are of fundamental importance for the efficient training of machine learning (ML) models. Activ
Externí odkaz:
http://arxiv.org/abs/2205.11117
Autor:
Arvanitidis, Georgios, González-Duque, Miguel, Pouplin, Alison, Kalatzis, Dimitris, Hauberg, Søren
Latent space geometry has shown itself to provide a rich and rigorous framework for interacting with the latent variables of deep generative models. The existing theory, however, relies on the decoder being a Gaussian distribution as its simple repar
Externí odkaz:
http://arxiv.org/abs/2106.05367
We present a framework for learning probability distributions on topologically non-trivial manifolds, utilizing normalizing flows. Current methods focus on manifolds that are homeomorphic to Euclidean space, enforce strong structural priors on the le
Externí odkaz:
http://arxiv.org/abs/2106.03500
Autor:
Sorteberg, Wilhelm E., Garasto, Stef, Pouplin, Alison S., Cantwell, Chris D., Bharath, Anil A.
Humans gain an implicit understanding of physical laws through observing and interacting with the world. Endowing an autonomous agent with an understanding of physical laws through experience and observation is seldom practical: we should seek altern
Externí odkaz:
http://arxiv.org/abs/1812.01609
We propose a novel deep learning model for classifying medical images in the setting where there is a large amount of unlabelled medical data available, but labelled data is in limited supply. We consider the specific case of classifying skin lesions
Externí odkaz:
http://arxiv.org/abs/1801.00693