AdaWeather: Adaptively Mixing Probabilistic Weather Forecasts with Logarithmic Regret
For weather forecasting, AdaWeather provides a principled adaptive combination method with theoretical guarantees, improving upon existing approaches that lack such regret bounds.
AdaWeather adaptively combines multiple probabilistic weather forecasts, achieving logarithmic regret relative to the best static mixture of experts in hindsight, and shows empirical improvements in temperature forecasting.
Recent advances in machine learning have produced probabilistic weather forecasting models comparable to state-of-the-art numerical weather predictors. But no model consistently dominates spatio-temporally, and relative performance is highly context-dependent. This motivates adaptive methods for combining multiple forecasts to obtain improvements and robustness. While combined forecasts have been proposed in the literature, these are achieved either through supervised learning or through prediction with expert advice methods. We introduce AdaWeather, an adaptive framework that combines many probabilistic forecasts using both machine learning as well as mixture of experts to arrive at a unified improved probabilistic forecast. While traditional expert methods develop the regret bounds with respect to the best single expert in hindsight, we extend the algorithm and analysis to show our method has logarithmic regret compared to the best static mixture of experts in hindsight. Empirically, we focus on forecasting temperature, and observe improvements over existing methods.