Metraipolis › archive › dc_cec85e536598

When Temporal Perturbations Act Like Sensor Biases: Label-Free Auditing of Wearable Activity Recognizers — arXiv:2609.29937

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.29937
License (as archived)
CC0-1.0
Archived
2026-10-08T04:27:52.678Z
SHA-256 of the archived text
0fa413f330f35fe2fa38c66512ea83456e773fa35e12cd6d1a548d357285a9db
Stable URL of this page
https://metraipolis.com/documento/dc_cec85e536598

How to cite

When Temporal Perturbations Act Like Sensor Biases: Label-Free Auditing of Wearable Activity Recognizers — arXiv:2609.29937 — arXiv (en), https://arxiv.org/abs/2609.29937 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cec85e536598 (Metraipolis archive dc_cec85e536598, sha256 0fa413f330f35fe2fa38c66512ea83456e773fa35e12cd6d1a548d357285a9db)
BibTeX
@misc{dccec85e536598,
  title = {When Temporal Perturbations Act Like Sensor Biases: Label-Free Auditing of Wearable Activity Recognizers — arXiv:2609.29937},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29937},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cec85e536598; sha256 0fa413f330f35fe2fa38c66512ea83456e773fa35e12cd6d1a548d357285a9db},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29937. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Wearable human-activity recognition (HAR) models operate across sensors, subjects, and backbones, yet a smooth waveform may appear temporal while exploiting a persistent sensor offset primarily. We introduce SpectrumAudit, a label-sealed audit that fits a phase-randomized full-window stimulus on calibration windows from subjects held out from training and testing. After selection, it replays its exact DC projection and budget-constrained zero-mean residual on the same frozen victim without refitting. Across 27 victims from three datasets and three backbones, the selected waveforms cause 2.87-40.83-point three-phase robust accuracy losses. Under this replay budget, DC is more damaging than AC on 24/27 victims and recovers at least 90% of the full drop on 22/27; all 5 failures occur on WISDM. In a held-out UTD-MHAD check, the selected waveform causes 13.49-pp accuracy and 11.68-pp macro-F1 losses, versus -0.66 pp for matched random changes. The audit diagnoses offset versus zero-mean variation under a common peak-budget cap. The code will be released upon acceptance.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2609.29937 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T03:50:58.188Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).