A cross-validated prediction-powered calibration technique that fine-tunes predictors and estimates synthetic label bias to produce prediction sets with coverage guarantees for wireless indoor localization using scarce calibration data.
Semi-supervised learning via cross- prediction-powered inference for wireless systems,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Reliable Wireless Indoor Localization via Cross-Validated Prediction-Powered Calibration
A cross-validated prediction-powered calibration technique that fine-tunes predictors and estimates synthetic label bias to produce prediction sets with coverage guarantees for wireless indoor localization using scarce calibration data.