Benutzer: Gast  Login
Titel:

Measuring Robustness in Deep Learning Based Compressive Sensing

Dokumenttyp:
Konferenzbeitrag
Autor(en):
Darestani, Mohammad Zalbagi; Chaudhari, Akshay S; Heckel, Reinhard
Abstract:
Deep neural networks give state-of-the-art accuracy for reconstructing images from few and noisy measurements, a problem arising for example in accelerated magnetic resonance imaging (MRI). However, recent works have raised concerns that deep-learning-based image reconstruction methods are sensitive to perturbations and are less robust than traditional methods: Neural networks (i) may be sensitive to small, yet adversarially-selected perturbations, (ii) may perform poorly under distribution shif...     »
Herausgeber:
Meila, Marina; Zhang, Tong
Kongress- / Buchtitel:
Proceedings of the 38th International Conference on Machine Learning
Band / Teilband / Volume:
139
Verlag / Institution:
PMLR
Jahr:
2021
Monat:
18--24 Jul
Seiten:
2433--2444
Serientitel:
Proceedings of Machine Learning Research
WWW:
https://proceedings.mlr.press/v139/darestani21a.html
 BibTeX