Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Online conformal prediction with decaying step sizes

About

We introduce a method for online conformal prediction with decaying step sizes. Like previous methods, ours possesses a retrospective guarantee of coverage for arbitrary sequences. However, unlike previous methods, we can simultaneously estimate a population quantile when it exists. Our theory and experiments indicate substantially improved practical properties: in particular, when the distribution is stable, the coverage is close to the desired level for every time point, not just on average over the observed sequence.

Anastasios N. Angelopoulos, Rina Foygel Barber, Stephen Bates• 2024

Related benchmarks

TaskDatasetResultRank
Conformal PredictionChangepoint simulated (test)
Coverage90
24
Post-shift coverage recoveryChangepoint Shift 1 (Time 1)
Recovery Time6
24
Time-series interval forecastingGoogle Stock
Coverage90.7
24
Post-shift coverage recoveryChangepoint Shift 2 (Time 2)
Recovery Time0.00e+0
24
Time-series interval forecastingTemperature
Coverage90.4
24
Online Conformal PredictionHeavy-tailed
Coverage90.8
24
Conformal PredictionDistribution Drift simulated (test)
Coverage90.6
24
Time-series interval forecastingElectricity Demand
Coverage90.1
24
Online Conformal PredictionVariance Changepoint
Coverage0.902
24
Time-series interval forecastingAmazon Stock
Coverage89.9
24
Showing 10 of 12 rows

Other info

Follow for update