Tiny, always-on and fragile: Bias propagation through design choices in on-device machine learning workflows. (arXiv:2201.07677v1 [cs.LG])

Billions of distributed, heterogeneous and resource constrained smart
consumer devices deploy on-device machine learning (ML) to deliver private,
fast and offline inference on personal data. On-device ML systems are highly
context dependent, and sensitive to user, usage, hardware and environmental
attributes. Despite this sensitivity and the propensity towards bias in ML,
bias in on-device ML has not been studied. This paper studies the propagation
of bias through design choices in on-device ML development workflows. We
position \emph{reliablity bias}, which arises …

arxiv bias design learning machine machine learning