Given a training set of input-output pairs (x_i,y_i), the method automatically learns a predictor of y given a new input x. The predictor is updated online for each observed input output pair.
For usage, please cite: D. Zachariah, P. Stoica and T.B. Schön "Online Learning for Distribution-Free Prediction" https://arxiv.org/abs/1703.05060, 2017