Conformalised Conditional Normalising Flows for Joint Prediction Regions in time series
This work addresses uncertainty quantification for time series forecasting, offering a domain-specific improvement for probabilistic models.
The paper tackled the challenge of applying conformal prediction to probabilistic generative models like normalising flows for multi-step time series forecasting, resulting in a method that generates disjoint prediction regions to improve predictive efficiency in multimodal distributions.
Conformal Prediction offers a powerful framework for quantifying uncertainty in machine learning models, enabling the construction of prediction sets with finite-sample validity guarantees. While easily adaptable to non-probabilistic models, applying conformal prediction to probabilistic generative models, such as Normalising Flows is not straightforward. This work proposes a novel method to conformalise conditional normalising flows, specifically addressing the problem of obtaining prediction regions for multi-step time series forecasting. Our approach leverages the flexibility of normalising flows to generate potentially disjoint prediction regions, leading to improved predictive efficiency in the presence of potential multimodal predictive distributions.