An aggregating strategy for shifting experts in discrete sequence prediction
This addresses the challenge of dynamic expert selection in sequence prediction for applications like time-series forecasting, though it appears incremental as it builds on existing exponential weighting methods.
The paper tackles the problem of adapting predictors to non-stationary environments with multiple expert advice, where the best expert changes over time. It proposes an algorithm based on exponential weighting with discounting, providing theoretical regret bounds and numerical verification on real-life datasets.
We study how we can adapt a predictor to a non-stationary environment with advises from multiple experts. We study the problem under complete feedback when the best expert changes over time from a decision theoretic point of view. Proposed algorithm is based on popular exponential weighing method with exponential discounting. We provide theoretical results bounding regret under the exponential discounting setting. Upper bound on regret is derived for finite time horizon problem. Numerical verification of different real life datasets are provided to show the utility of proposed algorithm.