Citation link:
https://doi.org/10.26092/elib/298
Double Backpropagation with Applications to Robustness and Saliency Map Interpretability
File | Description | Size | Format | |
---|---|---|---|---|
00108645-1.pdf | 22.8 MB | Adobe PDF | View/Open |
Other Titles: | Doppelte Fehler-Rückpropagierung mit Anwendungen zu Robustheit und Interpretierbarkeit der Salienzkarte | Authors: | Etmann, Christian | Supervisor: | Maaß, Peter | 1. Expert: | Maaß, Peter | Experts: | Möller, Michael | Abstract: | This thesis is concerned with works in connection to double backpropagation, which is a phenomenon that arises when first-order optimization methods are applied to a neural network's loss function, if this contains derivatives. Its connection to robustness and saliency map interpretability is explained. |
Keywords: | deep learning; neural networks; artificial intelligence; double backpropagation; robustness; saliency map; interpretability; computer vision; MALDI; mass spectrometry; machine learning; data science; statistics | Issue Date: | 14-Feb-2020 | Type: | Dissertation | Secondary publication: | no | DOI: | 10.26092/elib/298 | URN: | urn:nbn:de:gbv:46-00108645-13 urn:nbn:de:gbv:46-elib45018 |
Institution: | Universität Bremen | Faculty: | Fachbereich 03: Mathematik/Informatik (FB 03) |
Appears in Collections: | Dissertationen |
Page view(s)
227
checked on Apr 18, 2024
Download(s)
286
checked on Apr 18, 2024
Google ScholarTM
Check
Items in Media are protected by copyright, with all rights reserved, unless otherwise indicated.