Volgen
Erik Englesson
Erik Englesson
PhD student, KTH
Geverifieerd e-mailadres voor kth.se - Homepage
Titel
Geciteerd door
Geciteerd door
Jaar
Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labels
E Englesson, H Azizpour
Advances in Neural Information Processing Systems 34, 30284-30297, 2021
812021
Consistency Regularization can Improve Robustness to Label Noise
E Englesson, H Azizpour
International Conference on Machine Learning, UDL Workshop, 2021
182021
Efficient Evaluation-Time Uncertainty Estimation by Improved Distillation
E Englesson, H Azizpour
International Conference on Machine Learning, UDL Workshop, 2019
92019
Deep Double Descent via Smooth Interpolation
M Gamba, E Englesson, M Björkman, H Azizpour
Transactions on Machine Learning Research, 2023
82023
Robust Classification via Regression for Learning with Noisy Labels
E Englesson, H Azizpour
The Twelfth International Conference on Learning Representations, 2023
12023
Logistic-Normal Likelihoods for Heteroscedastic Label Noise
E Englesson, A Mehrpanah, H Azizpour
Transactions on Machine Learning Research, 2023
12023
Indirectly Parameterized Concrete Autoencoders
A Nilsson, K Wijk, E Englesson, A Hotti, C Saccardi, O Kviman, ...
arXiv preprint arXiv:2403.00563, 2024
2024
On Label Noise in Image Classification: An Aleatoric Uncertainty Perspective
E Englesson
KTH Royal Institute of Technology, 2024
2024
Het systeem kan de bewerking nu niet uitvoeren. Probeer het later opnieuw.
Artikelen 1–8