NEWMA: a new method for scalable model-free online change-point detection
This addresses the need for scalable change-point detection in real-time data streams, offering a practical solution for applications requiring low memory and computational overhead.
The paper tackles the problem of detecting abrupt changes in multi-dimensional time series distributions with limited computing resources, proposing NEWMA, a model-free online method that uses two EWMA statistics with different forgetting factors and Random Features for efficient distance computation, resulting in significantly faster detection than non-parametric methods for a given accuracy.
We consider the problem of detecting abrupt changes in the distribution of a multi-dimensional time series, with limited computing power and memory. In this paper, we propose a new, simple method for model-free online change-point detection that relies only on fast and light recursive statistics, inspired by the classical Exponential Weighted Moving Average algorithm (EWMA). The proposed idea is to compute two EWMA statistics on the stream of data with different forgetting factors, and to compare them. By doing so, we show that we implicitly compare recent samples with older ones, without the need to explicitly store them. Additionally, we leverage Random Features (RFs) to efficiently use the Maximum Mean Discrepancy as a distance between distributions, furthermore exploiting recent optical hardware to compute high-dimensional RFs in near constant time. We show that our method is significantly faster than usual non-parametric methods for a given accuracy.