EE
Erik Englesson
cs.LGstat.MLcs.CVcs.AI
On Valency
published · living versionsW_p626bb6d·v1 · currentpublished
Forgetting to Improve: Principled Data Removal in Active Learning
with Manuel Wendl, Andreas Krause, Carl Henrik Ek
1 version
Preprints & journals
8 papers in the corpus · 2019–2025On Spectral Properties of Gradient-based Explanation Methods2508.10595v1 · Amir Mehrpanah, Erik Englesson, Hossein Azizpour2025 · 1 citationECCV 2024, LNCS 15145, pp. 282-299
Medical Image Segmentation with SAM-generated Annotations2409.20253v1 · Iira Hakkinen, Iaroslav Melekhov, Erik Englesson et al.2024 · 8 citationsarXiv
Indirectly Parameterized Concrete Autoencoders2403.00563v2 · Alfred Nilsson, Klas Wijk, Sai bharath chandra Gutha et al.2024 · 1 citationarXiv
Logistic-Normal Likelihoods for Heteroscedastic Label Noise2304.02849v2 · Erik Englesson, Amir Mehrpanah, Hossein Azizpour2023 · 0 citationsarXiv
Deep Double Descent via Smooth Interpolation2209.10080v4 · Matteo Gamba, Erik Englesson, M\aarten Bjorkman et al.2022 · 1 citationarXiv
Consistency Regularization Can Improve Robustness to Label Noise2110.01242v1 · Erik Englesson, Hossein Azizpour2021 · 8 citationsarXiv
Generalized Jensen-Shannon Divergence Loss for Learning with Noisy Labels2105.04522v4 · Erik Englesson, Hossein Azizpour2021 · 69 citationsarXiv
Efficient Evaluation-Time Uncertainty Estimation by Improved Distillation1906.05419v1 · Erik Englesson, Hossein Azizpour2019 · 6 citationsarXiv
Career total: 10 works. 8 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.