Dynamic Local Regret for Non-convex Online Forecasting
This addresses forecasting challenges for machine learning practitioners dealing with concept drift and non-convexity, though it is incremental as it builds on existing regret frameworks.
The paper tackles online forecasting for non-convex models in dynamic environments by introducing a local regret framework, resulting in sublinear regret and improved stability, robustness, and efficiency compared to state-of-the-art methods on a real-world dataset.
We consider online forecasting problems for non-convex machine learning models. Forecasting introduces several challenges such as (i) frequent updates are necessary to deal with concept drift issues since the dynamics of the environment change over time, and (ii) the state of the art models are non-convex models. We address these challenges with a novel regret framework. Standard regret measures commonly do not consider both dynamic environment and non-convex models. We introduce a local regret for non-convex models in a dynamic environment. We present an update rule incurring a cost, according to our proposed local regret, which is sublinear in time T. Our update uses time-smoothed gradients. Using a real-world dataset we show that our time-smoothed approach yields several benefits when compared with state-of-the-art competitors: results are more stable against new data; training is more robust to hyperparameter selection; and our approach is more computationally efficient than the alternatives.