Reproduce “Reliable Wireless Indoor Localization via Cross-Validated Prediction-Powered Calibration” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Seonghoon Yoo, Houssem Sifaou, Sangwoo Park, Joonhyuk Kang, Osvaldo Simeone
RCPS-CPPI uses cross-validation to create risk-controlling location sets from Wi-Fi RSSI, letting labeled examples train pseudo-labeling predictors and correct their bias. It combines corrections with unlabeled readings to select sets under a risk bound. In one fingerprinting experiment, with 50 labeled calibration examples, average set radius was about 30% lower than conventional RCPS; this is not an accuracy or runtime result.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.