Application of deep neural networks to medical imaging tasks has in some sense become commonplace. Still, a “thorn in the side” of the deep learning movement is the argument that deep networks are prone to overfitting and are thus unable to generalize well when datasets are small (as is common in medical imaging tasks). One way to bolster confidence is to provide mathematical guarantees, or bounds, on network performance after training which explicitly quantify the possibility of overfitting. In this work, we explore recent advances using the PAC-Bayesian framework to provide bounds on generalization error for large (stochastic) networks. While previous efforts focus on classification in larger natural image datasets (e.g., MNIST and CIFAR-10), we apply these techniques to both classification and segmentation in a smaller medical imagining dataset: the ISIC 2018 challenge set. We observe the resultant bounds are competitive compared to a simpler baseline, while also being more explainable and alleviating the need for holdout sets.
|Title of host publication||Medical Image Computing and Computer Assisted Intervention – MICCAI 2021 - 24th International Conference, Proceedings|
|Editors||Marleen de Bruijne, Philippe C. Cattin, Stéphane Cotin, Nicolas Padoy, Stefanie Speidel, Yefeng Zheng, Caroline Essert|
|Publisher||Springer Science and Business Media Deutschland GmbH|
|Number of pages||11|
|Publication status||Published - 2021|
|Event||24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021 - Virtual, Online|
Duration: 2021 Sep 27 → 2021 Oct 1
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021|
|Period||21/9/27 → 21/10/1|
Bibliographical noteFunding Information:
Acknowledgment. This work is supported by the University of Pittsburgh Alzheimer Disease Research Center Grant (P30 AG066468).
© 2021, Springer Nature Switzerland AG.
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)