2015 (modified: 11 Nov 2022)ICML 2015Readers: Everyone
Abstract:We consider the problem of time series prediction in the presence of missing data. We cast the problem as an online learning problem in which the goal of the learner is to minimize prediction error...