Reproduce “Conformal Prediction for Time Series with Deep Sequence Models” 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
Junghwan Lee, Jonghyeok Lee, Yao Xie
The authors study three ways to use deep sequence models for time-series conformal intervals: quantile heads, quantile-level-conditioned functions, and representation-based local weighting. Encoded recent context helps estimate residual quantiles or weight calibration errors. On Solar data with Chronos, DSCP-Local and ResCP have the same coverage gap (+0.011), while DSCP-Local has a lower Winkler score (109.9 versus 130.3).
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.