Zobrazeno 1 - 10
of 976
pro vyhledávání: '"Lomuscio, A."'
Autor:
Brückner, Benedikt, Lomuscio, Alessio
We develop a method for the efficient verification of neural networks against convolutional perturbations such as blurring or sharpening. To define input perturbations we use well-known camera shake, box blur and sharpen kernels. We demonstrate that
Externí odkaz:
http://arxiv.org/abs/2411.04594
We address the problem of verifying neural networks against geometric transformations of the input image, including rotation, scaling, shearing, and translation. The proposed method computes provably sound piecewise linear constraints for the pixel v
Externí odkaz:
http://arxiv.org/abs/2408.13140
Autor:
Walker, Thomas, Lomuscio, Alessio
Probably Approximately Correct (PAC) bounds are widely used to derive probabilistic guarantees for the generalisation of machine learning models. They highlight the components of the model which contribute to its generalisation capacity. However, cur
Externí odkaz:
http://arxiv.org/abs/2407.20122
Autor:
Prot, Aubin JC. M., Melchiorre, Michele, Schaaf, Tilly, Poeira, Ricardo G., Elanzeery, Hossam, Lomuscio, Alberto, Oueslati, Souhaib, Zelenina, Anastasia, Dalibor, Thomas, Kusch, Gunnar, Hu, Yucheng, Oliver, Rachel A., Siebentritt, Susanne
Alloying small quantities of silver into Cu(In,Ga)Se2 was shown to improve the efficiency for wide and low band gap solar cells. We study low band gap industrial Cu(In,Ga)(S,Se)2 absorbers, substituting less than 10% of the copper with silver, using
Externí odkaz:
http://arxiv.org/abs/2403.04394
We introduce two algorithms for computing tight guarantees on the probabilistic robustness of Bayesian Neural Networks (BNNs). Computing robustness guarantees for BNNs is a significantly more challenging task than verifying the robustness of standard
Externí odkaz:
http://arxiv.org/abs/2401.11627
Parameterized Kerr spacetimes allow us to test the nature of black holes in model-independent ways. Such spacetimes contain several arbitrary functions and, as a matter of practicality, one Taylor expands them about infinity and keeps only to finite
Externí odkaz:
http://arxiv.org/abs/2311.08659
Autor:
Prot, Aubin JC. M., Melchiorre, Michele, Dingwell, Felix, Zelenina, Anastasia, Elanzeery, Hossam, Lomuscio, Alberto, Dalibor, Thomas, Guc, Maxim, Fonoll-Rubio, Robert, Izquierdo-Roca, Victor, Kusch, Gunnar, Oliver, Rachel A., Siebentritt, Susanne
Record efficiency in chalcopyrite-based solar cells Cu(In,Ga)(S,Se)2 is achieved using a gallium gradient to increase the band gap of the absorber towards the back side. Although this structure has successfully reduced recombination at the back conta
Externí odkaz:
http://arxiv.org/abs/2307.02356
Autor:
De Palma, Alessandro, Bunel, Rudy, Dvijotham, Krishnamurthy, Kumar, M. Pawan, Stanforth, Robert, Lomuscio, Alessio
In order to train networks for verified adversarial robustness, it is common to over-approximate the worst-case loss over perturbation regions, resulting in networks that attain verifiability at the expense of standard performance. As shown in recent
Externí odkaz:
http://arxiv.org/abs/2305.13991
Autor:
To, Andy S. H., James, Alexander W., Bastian, T. S., van Driel-Gesztelyi, Lidia, Long, David M., Baker, Deborah, Brooks, David H., Lomuscio, Samantha, Stansby, David, Valori, Gherardo
Sun-as-a-star coronal plasma composition, derived from full-Sun spectra, and the F10.7 radio flux (2.8 GHz) have been shown to be highly correlated (r = 0.88) during solar cycle 24. However, this correlation becomes nonlinear during increased solar m
Externí odkaz:
http://arxiv.org/abs/2304.02552
Publikováno v:
2022 IEEE 9th International Conference on Data Science and Advanced Analytics (DSAA), Shenzhen, China, 2022, pp. 1-10
Several explainable AI methods allow a Machine Learning user to get insights on the classification process of a black-box model in the form of local linear explanations. With such information, the user can judge which features are locally relevant fo
Externí odkaz:
http://arxiv.org/abs/2302.07760