PuSH - Publikationsserver des Helmholtz Zentrums München

Fischer, S.M. ; Felsner, L. ; Osuala, R. ; Kiechle, J. ; Lang, D.M. ; Peeken, J.C. ; Schnabel, J.A.

Progressive growing of patch size: Resource-efficient curriculum learning for dense prediction tasks.

In: (Medical Image Computing and Computer Assisted Intervention – MICCAI 2024). Berlin [u.a.]: Springer, 2024. 510-520 (Lect. Notes Comput. Sc. ; 15009 LNCS)
DOI
In this work, we introduce Progressive Growing of Patch Size, a resource-efficient implicit curriculum learning approach for dense prediction tasks. Our curriculum approach is defined by growing the patch size during model training, which gradually increases the task’s difficulty. We integrated our curriculum into the nnU-Net framework and evaluated the methodology on all 10 tasks of the Medical Segmentation Decathlon. With our approach, we are able to substantially reduce runtime, computational costs, and CO2 emissions of network training compared to classical constant patch size training. In our experiments, the curriculum approach resulted in improved convergence. We are able to outperform standard nnU-Net training, which is trained with constant patch size, in terms of Dice Score on 7 out of 10 MSD tasks while only spending roughly 50% of the original training runtime. To the best of our knowledge, our Progressive Growing of Patch Size is the first successful employment of a sample-length curriculum in the form of patch size in the field of computer vision.
Altmetric
Weitere Metriken?
[➜Einloggen]
Zusatzinfos bearbeiten [➜Einloggen]
Publikationstyp Artikel: Konferenzbeitrag
Korrespondenzautor
Schlagwörter Curriculum Learning ; Medical Segmentation Decathlon ; Nnu-net ; Resource Efficiency ; Segmentation
ISSN (print) / ISBN 0302-9743
e-ISSN 1611-3349
Konferenztitel Medical Image Computing and Computer Assisted Intervention – MICCAI 2024
Zeitschrift Lecture Notes in Computer Science
Quellenangaben Band: 15009 LNCS, Heft: , Seiten: 510-520 Artikelnummer: , Supplement: ,
Verlag Springer
Verlagsort Berlin [u.a.]
Nichtpatentliteratur Publikationen
Institut(e) Institute for Machine Learning in Biomed Imaging (IML)
Institute of Radiation Medicine (IRM)