Problem
Research questions and friction points this paper is trying to address.
Reliability Routing
Wearable Stress Prediction
Low-Stakes Settings
Output Availability
Fixed-Label
ReliaGate: Reliability Routing for Low-Stakes Wearable Stress Prediction
🤖 AI Summary
This study addresses the insufficient aggregation accuracy of wearable stress prediction in low-risk scenarios by proposing a Fixed-Label Reliability Routing framework. The method constructs a posterior correctness scoring system integrating confidence, signal quality, and geometric cues, employing a gating mechanism to dynamically decide whether to output or retain labels without modifying original annotations, thereby optimizing error distribution. Validations on the WESAD and UBFC-Phys datasets demonstrate that this framework effectively balances surface label errors against output availability. Consequently, it significantly enhances both the reliability and practical utility of prediction results, offering a robust solution for improving wearable sensing applications where maintaining data integrity while ensuring actionable outputs is critical.
📝 Abstract
We study when a wearable stress system should surface a prediction rather than change it. In low-stakes reflection and summary settings, aggregate accuracy is insufficient because withholding can reduce error while leaving some people with little or no information. We formulate fixed-label reliability routing: after a locked classifier emits a protocol-defined stress/non-stress label, a post-hoc gate surfaces that unchanged label or withholds it as unavailable. ReliaGate assembles established confidence, signal-quality/trust, agreement, train-standardized atypicality, and train-fitted geometry cues into a post-hoc correctness score. We evaluate four wearable datasets using subject-disjoint folds, validation-selected routing, paired held-out-subject intervals, and pooled and per-subject analyses. WESAD point estimates favored ReliaGate, UBFC-Phys primary coverage/risk intervals favored ReliaGate, and E4 checks were mixed. ReliaGate provides an operational framework for studying surfaced-label error, output availability, and accepted-output distribution across subjects, without revising labels or providing clinical or finite-sample risk guarantees.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Reliability Routing
Post-hoc Gate
Fixed-label
Wearable Stress Prediction
Correctness Score
🔎 Similar Papers
No similar papers found.
