Folgen
Eugenio Clerico
Eugenio Clerico
Department of Statistics, University of Oxford
Bestätigte E-Mail-Adresse bei stats.ox.ac.uk
Titel
Zitiert von
Zitiert von
Jahr
Stable resnet
S Hayou, E Clerico, B He, G Deligiannidis, A Doucet, J Rousseau
International Conference on Artificial Intelligence and Statistics, 1324-1332, 2021
512021
Conditionally Gaussian PAC-Bayes
E Clerico, G Deligiannidis, A Doucet
The 25th International Conference on Artificial Intelligence and Statistics, 2022
192022
Chained Generalisation Bounds
E Clerico, A Shidani, G Deligiannidis, A Doucet
35th Annual Conference on Learning Theory (COLT 2022) 178, 4212-4257, 2022
172022
Wide stochastic networks: Gaussian limit and PAC-Bayesian training
E Clerico, G Deligiannidis, A Doucet
International Conference on Algorithmic Learning Theory, 447-470, 2023
102023
Generalisation under gradient descent via deterministic PAC-Bayes
E Clerico, T Farghly, G Deligiannidis, B Guedj, A Doucet
arXiv preprint arXiv:2209.02525, 2023
7*2023
Generalization bounds for mixing processes via delayed online-to-PAC conversions
B Abeles, E Clerico, G Neu
arXiv preprint arXiv:2406.12600, 2024
2024
A note on regularised NTK dynamics with an application to PAC-Bayesian training
E Clerico, B Guedj
arXiv preprint arXiv:2312.13259, 2023
2023
Generalisation and expressiveness for over-parameterised neural networks
E Clerico
University of Oxford, 2023
2023
Das System kann den Vorgang jetzt nicht ausführen. Versuchen Sie es später erneut.
Artikel 1–8