ECCE Seminar - Stochastic Online Convex Optimization with Applications to Adaptive Forecasting

Please use this Zoom link (Meeting ID: 652 577 1125, Password: Wintenberg) for online attendance.

Stochastic Online Convex Optimization with Applications to Adaptive Forecasting 

We propose a general framework for stochastic online convex optimization. Certain algorithms, including online Newton steps and a scale-free variant of Bernstein online aggregation, achieve the best-known rates in unbounded stochastic settings. We apply it to calibrating parametric probabilistic forecasters of non-stationary sub-Gaussian time series. We illustrate the benefit of our approach to real-world data such as electricity load and temperatures.