Handling temporal variation of unknown characteristics in streaming data analysis.

Опубликовано 17 августа 2016, 22:31

Data collection technology is undergoing a revolution that is enabling streaming acquisition of real-time information in a wide variety of settings. Faced with indefinitely long, high frequency and possibly high dimensional data sequences, learning algorithms must rely on summary statistics and computationally efficient online inference without the need to store and revisit the data history. Moreover, learning must be temporally adaptive in order to remain up-to-date against unforeseen changes, smooth or abrupt, in the underlying data generation mechanism. In cases where explicit dynamic modelling is either impossible or impractical, temporally adaptive behaviour may still be induced by controlling the responsiveness of the estimator to the novel information. We discuss ways in which this can be accomplished in data-dependent manners for popular classes of online algorithms. We focus on the Robbins-Monro family of algorithms that naturally feature a sequence of user-specified learning rates, and discuss available methodology for automatic self-tuning in this context. On the basis of both theoretical insights and real-data experiments, we demonstrate that such approaches can efficiently handle temporal variation of unknown characteristics, while additionally serving as a monitoring tool.