mlsanityData to 7 Oct 2026

SNAP-UQ: Self-supervised Next-Activation Prediction for Single-Pass Uncertainty in TinyML

arXiv 2508.12907 ↗First posted 18 Aug 2025cs.LG cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Reliable uncertainty estimation is a key missing piece for on-device monitoring in TinyML: microcontrollers must detect failures, distribution shift, or accuracy drops under strict flash/latency budgets, yet common uncertainty approaches (deep ensembles, MC dropout, early exits, temporal buffering) typically require multiple passes, extra branches, or state that is impractical on milliwatt hardware. This paper proposes a novel and practical method, SNAP-UQ, for single-pass, label-free uncertainty estimation based on depth-wise next-activation prediction. SNAP-UQ taps a small set of backbone layers and uses tiny int8 heads to predict the mean and scale of the next activation from a low-rank projection of the previous one; the resulting standardized prediction error forms a depth-wise surprisal signal that is aggregated and mapped through a lightweight monotone calibrator into an actionable uncertainty score. The design introduces no temporal buffers or auxiliary exits and preserves state-free inference, while increasing deployment footprint by only a few tens of kilobytes. Across vision and audio backbones, SNAP-UQ reduces flash and latency relative to early-exit and deep-ensemble baselines (typically ~40--60% smaller and ~25--35% faster), with several competing methods at similar accuracy often exceeding MCU memory limits. On corrupted streams, it improves accuracy-drop event detection by multiple AUPRC points and maintains strong failure detection (AUROC ≈ 0.9) in a single forward pass. By grounding uncertainty in layer-to-layer dynamics rather than solely in output confidence, SNAP-UQ offers a novel, resource-efficient basis for robust TinyML monitoring. Our code is available at: https://github.com/Ism-ail11/SNAP-UQ

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 42 of its 61 references that are in the map. First posted on arXiv 18 Aug 2025.

Builds on

42 of 61 references.

References · 42

  1. 2017On Calibration of Modern Neural Networkscited by 4,860
  2. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  3. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  4. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  5. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,827
  6. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  7. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  8. 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
  9. 2018A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attackscited by 1,472
  10. 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,325
  11. 2020Energy-based Out-of-distribution Detectioncited by 1,296
  12. 2019Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shiftcited by 1,274
30 more
  1. 2019AugMix: A Simple Data Processing Method to Improve Robustness and Uncertaintycited by 897
  2. 2019Once-for-All: Train One Network and Specialize it for Efficient Deploymentcited by 778
  3. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 741
  4. 2018Evidential Deep Learning to Quantify Classification Uncertaintycited by 728
  5. 2019Learned Step Size Quantizationcited by 604
  6. 2020Up or Down? Adaptive Rounding for Post-Training Quantizationcited by 600
  7. 2017BranchyNet: Fast Inference via Early Exiting from Deep Neural Networkscited by 591
  8. 2018Predictive Uncertainty Estimation via Prior Networkscited by 555
  9. 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
  10. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 464
  11. 2018PACT: Parameterized Clipping Activation for Quantized Neural Networkscited by 456
  12. 2020CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instancescited by 441
  13. 2020Generalized ODIN: Detecting Out-of-distribution Image without Learning from Out-of-distribution Datacited by 408
  14. 2022ViM: Out-Of-Distribution with Virtual-logit Matchingcited by 404
  15. 2021Revisiting the Calibration of Modern Neural Networkscited by 387
  16. 2020Uncertainty Sets for Image Classifiers using Conformal Predictioncited by 365
  17. 2019Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibrationcited by 359
  18. 2022VOS: Learning What You Don't Know by Virtual Outlier Synthesiscited by 300
  19. 2022Extremely Simple Activation Shaping for Out-of-Distribution Detectioncited by 212
  20. 2021Deep Deterministic Uncertainty: A Simple Baselinecited by 200
  21. 2020Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Countscited by 163
  22. 2020TensorFlow Lite Micro: Embedded Machine Learning on TinyML Systemscited by 161
  23. 2019MNIST-C: A Robustness Benchmark for Computer Visioncited by 137
  24. 2021MLPerf Tiny Benchmarkcited by 118
  25. 2018CMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUscited by 116
  26. 2017Hello Edge: Keyword Spotting on Microcontrollerscited by 108
  27. 2023Uncertainty Estimation by Fisher Information-based Evidential Deep Learningcited by 58
  28. 2023Fixing Overconfidence in Dynamic Neural Networkscited by 19
  29. 2024QUTE: Quantifying Uncertainty in TinyML with Early-exit-assisted ensembles for model-monitoringcited by 4
  30. 2024Tiny Deep Ensemble: Uncertainty Estimation in Edge AI Accelerators via Ensembling Normalization Layers with Shared Weightscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025TCUQ: Single-Pass Uncertainty Quantification from Temporal Consistency with Streaming Conformal Calibration for TinyML
  2. 2024UR2M: Uncertainty and Resource-Aware Event Detection on Microcontrollers
  3. 2025On-Sensor Convolutional Neural Networks with Early-Exits
  4. 2024Exploring the Boundaries of On-Device Inference: When Tiny Falls Short, Go Hierarchical
  5. 2025Hardware-efficient tractable probabilistic inference for TinyML Neurosymbolic AI applications
  6. 2025Tiny but Mighty: A Software-Hardware Co-Design Approach for Efficient Multimodal Inference on Battery-Powered Small Devices
  7. 2025HARMONY: Hidden Activation Representations and Model Output-Aware Uncertainty Estimation for Vision-Language Models
  8. 2026VisionPsy-Nano: Improving Accuracy, Efficiency, and Reliability in On-Device Vision-Language Models