REVIEW 3 cited by
Online Multivalid Learning: Means, Moments, and Prediction Intervals
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Online Multivalid Learning: Means, Moments, and Prediction Intervals
read the original abstract
We present a general, efficient technique for providing contextual predictions that are "multivalid" in various senses, against an online sequence of adversarially chosen examples $(x,y)$. This means that the resulting estimates correctly predict various statistics of the labels $y$ not just marginally -- as averaged over the sequence of examples -- but also conditionally on $x \in G$ for any $G$ belonging to an arbitrary intersecting collection of groups $\mathcal{G}$. We provide three instantiations of this framework. The first is mean prediction, which corresponds to an online algorithm satisfying the notion of multicalibration from Hebert-Johnson et al. The second is variance and higher moment prediction, which corresponds to an online algorithm satisfying the notion of mean-conditioned moment multicalibration from Jung et al. Finally, we define a new notion of prediction interval multivalidity, and give an algorithm for finding prediction intervals which satisfy it. Because our algorithms handle adversarially chosen examples, they can equally well be used to predict statistics of the residuals of arbitrary point prediction methods, giving rise to very general techniques for quantifying the uncertainty of predictions of black box algorithms, even in an online adversarial setting. When instantiated for prediction intervals, this solves a similar problem as conformal prediction, but in an adversarial environment and with multivalidity guarantees stronger than simple marginal coverage guarantees.
Forward citations
Cited by 3 Pith papers
-
An Efficient Black-Box Reduction from Online Learning to Multicalibration, and a New Route to $\Phi$-Regret Minimization
Black-box reductions from no-regret online learning to multicalibration and from multicalibration to Phi-regret minimization are established, resolving the main open question in Garg et al. (SODA '24).
-
Multicalibration Boosting: Theory, Convergence, and Transferability
Unifies MCBoost variants and proves convergence to Bregman projection with rates, finite-sample bounds, and covariate-shift transferability.
-
Bi-Level Chaotic Fusion Based Graph Convolutional Network for Stock Market Prediction Interval
A spatio-temporal GCN with bi-level chaotic fusion and volatility-aware gating produces stock price prediction intervals on NSE data, reporting superior Winkler score, PIAW, and PICP over LSTM, GRU, GCN, and HGNN baselines.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.