On instabilities of deep learning in image reconstruction and the potential costs of AI

Deep learning, due to its unprecedented success in tasks such as image classification, has emerged as a new tool in image reconstruction with potential to change the field. In this paper, we demonstrate a crucial phenomenon: Deep learning typically yields unstable methods for image reconstruction. The instabilities usually occur in several forms: 1) Certain tiny, almost undetectable perturbations, both in the image and sampling domain, may result in severe artefacts in the reconstruction; 2) a small structural change, for example, a tumor, may not be captured in the reconstructed image; and 3) (a counterintuitive type of instability) more samples may yield poorer performance. Our stability test with algorithms…

• N
  Nvidia
• LT
  Leverhulme Trust
• RS
  Royal Society
• EC
  European Commission

  Awards: 655282, SFRH/BPD/118714/2016
• CN
  Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada

  Award: 611675
• NS
  Natural Sciences and Engineering Research Council of Canada
• EA
  Engineering and Physical Sciences Research Council

  Awards: EP/L003457/1, EP/L003457/1
• FP
  Fundação para a Ciência e a Tecnologia