CoChat
Sign Up
PreprintCited by 6Advanced Neural Network ApplicationsAdversarial Robustness in Machine LearningIoT and Edge/Fog Computing

The benefit of the doubt: Uncertainty aware sensing for edge computing platforms

Open access
Qendro, Lorena, Chauhan, Jagmohan, Ramos, Alberto Gil C. P., Mascolo, Cecilia·2021·arXiv
DOI 10.48550/arxiv.2102.05956arXiv 2102.05956
Open this paper in CoChat — free accountOpen PDF

Abstract

Neural networks (NNs) lack measures of "reliability" estimation that would enable reasoning over their predictions. Despite the vital importance, especially in areas of human well-being and health, state-of-the-art uncertainty estimation techniques are computationally expensive when applied to resource-constrained devices. We propose an efficient framework for predictive uncertainty estimation in NNs deployed on embedded edge systems with no need for fine-tuning or re-training strategies. To meet the energy and latency requirements of these embedded platforms the framework is built from the ground up to provide predictive uncertainty based only on one forward pass and a negligible amount of additional matrix multiplications with theoretically proven correctness. Our aim is to enable already trained deep learning models to generate uncertainty estimates on resource-limited devices at inference time focusing on classification tasks. This framework is founded on theoretical developments casting dropout training as approximate inference in Bayesian NNs. Our layerwise distribution approximation to the convolution layer cascades through the network, providing uncertainty estimates in one single run which ensures minimal overhead, especially compared with uncertainty techniques that require multiple forwards passes and an equal linear rise in energy and latency requirements making them unsuitable in practice. We demonstrate that it yields better performance and flexibility over previous work based on multilayer perceptrons to obtain uncertainty estimates. Our evaluation with mobile applications datasets shows that our approach not only obtains robust and accurate uncertainty estimations but also outperforms state-of-the-art methods in terms of systems performance, reducing energy consumption (up to 28x), keeping the memory overhead at a minimum while still improving accuracy (up to 16%).

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

DataCite · 10.48550/arxiv.2102.05956

Cites

54 of 54 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Bayesian Neural Networks impose prohibitive computational costs for MCUs, largely due to the overhead of representing weights as full distributions and the computationally intensive sampling or approximation techniques often required for inference.

    “While various methods exist for uncertainty quantification, many, like Bayesian Neural Networks (BNNs) [9], impose prohibitive computational costs [10], largely due to the overhead of representing weights as full distributions and the computationally intensive sampling or approximation techniques often required for inference.”

    From A cascade framework for on-device uncertainty-aware event detection on microcontrollers (2026) · cites this paper as [10]

    Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. [34][34] Seulki Lee and Shahriar Nirjon. 2019. Neuro. ZERO: a zero-energy neural network accelerator for embedded sensing and inference systems. In Proceedings of the 17th Conference on Embedded Networked Sensor Systems . ACM, 138-152. -
  2. [36][36] Hong Lu, Denise Frauendorfer, Mashfiqui Rabbi, Marianne Schmid Mast, Gokul T Chittaranjan, Andrew T Campbell, Daniel Gatica-Perez, and Tanzeem Choudhury. 2012. Stresssense: Detecting stress in unconstrained acoustic environments using smartphones. In Proceedings of the 2012 ACM Conference on Ubiquitous Computing . ACM, 351-360. -

Cited by

Sorted by citations

Related

Authors
Qendro, L., Chauhan, J., Ramos, A. G. C. P., & Mascolo, C.
ORCID
Venue
arXiv
Publisher
arXiv
Published
2021
ISSN
2331-8422 (ISSN-L)
DOI
10.48550/arxiv.2102.05956
arXiv
2102.05956
OpenAlex
W3128242499
Cited by
6
References
52
Citations by year
2025: 1 · 2021: 5
Type
posted content
Language
en
Open access
Licence
arxiv.org/licenses/nonexclusive-distrib/1.0/
Locations
  • arXiv (Cornell University) (submittedVersion)
  • Cambridge University Engineering Department Publications Database (submittedVersion)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
  • Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
Keywords
predictive uncertainty estimation, artificial neural networks, edge computing, resource-constrained devices, Bayesian neural networks, approximate inference, convolutional layers, energy efficiency, inference latency, mobile application, classification

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review