Change Point Detection by Cross-Entropy Maximization
This work addresses change point detection for data analysis, but it is incremental as it extends an existing framework with a new discrepancy measure.
The paper tackles offline unsupervised change point detection by proposing to maximize cross-entropy between successive segments instead of minimizing segment-wise costs, and demonstrates advantages over three state-of-the-art methods in experiments on two challenging datasets.
Many offline unsupervised change point detection algorithms rely on minimizing a penalized sum of segment-wise costs. We extend this framework by proposing to minimize a sum of discrepancies between segments. In particular, we propose to select the change points so as to maximize the cross-entropy between successive segments, balanced by a penalty for introducing new change points. We propose a dynamic programming algorithm to solve this problem and analyze its complexity. Experiments on two challenging datasets demonstrate the advantages of our method compared to three state-of-the-art approaches.