Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

JAPAN: Joint Adaptive Prediction Areas with Normalising-Flows

About

Conformal prediction provides a model-agnostic framework for uncertainty quantification with finite-sample validity guarantees, making it an attractive tool for constructing reliable prediction sets. However, existing approaches commonly rely on residual-based conformity scores, which impose geometric constraints and struggle when the underlying distribution is multimodal. In particular, they tend to produce overly conservative prediction areas centred around the mean, often failing to capture the true shape of complex predictive distributions. In this work, we introduce JAPAN (Joint Adaptive Prediction Areas with Normalising-Flows), a conformal prediction framework that uses density-based conformity scores. By leveraging flow-based models, JAPAN estimates the (predictive) density and constructs prediction areas by thresholding on the estimated density scores, enabling compact, potentially disjoint, and context-adaptive regions that retain finite-sample coverage guarantees. We theoretically motivate the efficiency of JAPAN and empirically validate it across multivariate regression and forecasting tasks, demonstrating good calibration and tighter prediction areas compared to existing baselines. We also provide several \emph{extensions} adding flexibility to our proposed framework.

Eshant English, Christoph Lippert• 2025

Related benchmarks

TaskDatasetResultRank
Prediction Region EstimationEnergy (test)
Coverage90.8
16
Conformal PredictionSpiral_L synthetic (test)
Coverage90.2
8
Conformal PredictionPinwheel_L synthetic (test)
Coverage90.4
8
Conformal PredictionNYC Taxi (20 random splits)
Coverage (%)90.2
8
Prediction Region EstimationRF1-2D (test)
Coverage90.8
8
Conformal PredictionSpiral_H synthetic (test)
Coverage90
8
Conformal PredictionPinwheel_H synthetic (test)
Coverage90.2
8
Prediction Region EstimationSCM20D (test)
Coverage90.2
8
Prediction Region EstimationRF1-4D (test)
Coverage (%)89.9
8
Showing 9 of 9 rows

Other info

Follow for update