Zobrazeno 1 - 10
of 25
pro vyhledávání: '"Miani, Marco"'
Bayesian deep learning all too often underfits so that the Bayesian prediction is less accurate than a simple point estimate. Uncertainty quantification then comes at the cost of accuracy. For linearized models, the null space of the generalized Gaus
Externí odkaz:
http://arxiv.org/abs/2410.16901
Current uncertainty quantification is memory and compute expensive, which hinders practical uptake. To counter, we develop Sketched Lanczos Uncertainty (SLU): an architecture-agnostic uncertainty score that can be applied to pre-trained neural networ
Externí odkaz:
http://arxiv.org/abs/2409.15008
Autor:
Roy, Hrittik, Miani, Marco, Ek, Carl Henrik, Hennig, Philipp, Pförtner, Marvin, Tatzel, Lukas, Hauberg, Søren
Current approximate posteriors in Bayesian neural networks (BNNs) exhibit a crucial limitation: they fail to maintain invariance under reparameterization, i.e. BNNs assign different posterior densities to different parametrizations of identical funct
Externí odkaz:
http://arxiv.org/abs/2406.03334
Autor:
Zepf, Kilian, Wanna, Selma, Miani, Marco, Moore, Juston, Frellsen, Jes, Hauberg, Søren, Warburg, Frederik, Feragen, Aasa
Image segmentation relies heavily on neural networks which are known to be overconfident, especially when making predictions on out-of-distribution (OOD) images. This is a common scenario in the medical domain due to variations in equipment, acquisit
Externí odkaz:
http://arxiv.org/abs/2303.13123
We propose the first Bayesian encoder for metric learning. Rather than relying on neural amortization as done in prior works, we learn a distribution over the network weights with the Laplace Approximation. We actualize this by first proving that the
Externí odkaz:
http://arxiv.org/abs/2302.01332
Autor:
Miani, Marco, Warburg, Frederik, Moreno-Muñoz, Pablo, Detlefsen, Nicke Skafte, Hauberg, Søren
Established methods for unsupervised representation learning such as variational autoencoders produce none or poorly calibrated uncertainty estimates making it difficult to evaluate if learned representations are stable and reliable. In this work, we
Externí odkaz:
http://arxiv.org/abs/2206.15078
Having access to an exploring restart distribution (the so-called wide coverage assumption) is critical with policy gradient methods. This is due to the fact that, while the objective function is insensitive to updates in unlikely states, the agent m
Externí odkaz:
http://arxiv.org/abs/2106.15503
Autor:
Pauletta, Margherita, Di Marco, Caterina, Frappa, Giada, Miani, Marco, Campione, Giuseppe, Russo, Gaetano
Publikováno v:
In Engineering Structures 1 February 2021 228
Autor:
Pauletta, Margherita, Di Marco, Caterina, Frappa, Giada, Somma, Giuliana, Pitacco, Igino, Miani, Marco, Das, Sreekanta, Russo, Gaetano
Publikováno v:
In Engineering Structures 1 December 2020 224
Publikováno v:
IEEE Transactions on Pattern Analysis and Machine Intelligence; December 2024, Vol. 46 Issue: 12 p11422-11431, 10p