Neural Conformal Control for Time Series Forecasting
This addresses the challenge of reliable uncertainty quantification in time series forecasting for applications like epidemics and energy demand, though it appears incremental as an enhancement to existing conformal prediction methods.
The paper tackles the problem of time series forecasting in non-stationary environments by introducing a neural network conformal prediction method that acts as a neural controller to achieve target coverage, leveraging multi-view data and monotonicity constraints. The method demonstrates significant improvements in coverage and probabilistic accuracy on real-world datasets from epidemics, electric demand, and weather, and is the only one combining good calibration with consistent prediction intervals.
We introduce a neural network conformal prediction method for time series that enhances adaptivity in non-stationary environments. Our approach acts as a neural controller designed to achieve desired target coverage, leveraging auxiliary multi-view data with neural network encoders in an end-to-end manner to further enhance adaptivity. Additionally, our model is designed to enhance the consistency of prediction intervals in different quantiles by integrating monotonicity constraints and leverages data from related tasks to boost few-shot learning performance. Using real-world datasets from epidemics, electric demand, weather, and others, we empirically demonstrate significant improvements in coverage and probabilistic accuracy, and find that our method is the only one that combines good calibration with consistency in prediction intervals.