Deep learning model solves change point detection for multiple change types

TL;DR

This paper introduces a deep learning approach for change point detection that effectively handles multiple distributions and semi-structured data, outperforming traditional classifiers in robustness and applicability to complex real-world scenarios.

Contribution

The paper presents a novel deep learning model capable of detecting change points across multiple distributions and semi-structured data, addressing limitations of existing binary-distribution assumptions.

Findings

Model outperforms classifiers in robustness

Effective on image sequence datasets

Handles multiple distributions in data

Abstract

A change points detection aims to catch an abrupt disorder in data distribution. Common approaches assume that there are only two fixed distributions for data: one before and another after a change point. Real-world data are richer than this assumption. There can be multiple different distributions before and after a change. We propose an approach that works in the multiple-distributions scenario. Our approach learn representations for semi-structured data suitable for change point detection, while a common classifiers-based approach fails. Moreover, our model is more robust, when predicting change points. The datasets used for benchmarking are sequences of images with and without change points in them.