InDiD: Instant Disorder Detection via Representation Learning

Abstract

Change points are abrupt alterations in the distribution of sequential data. A change-point detection (CPD) model aims at quick detection of such changes. Classic approaches perform poorly for semi-structured sequential data because of the absence of adequate data representation learning. To deal with it, we introduce a principled differentiable loss function that considers the specificity of the CPD task. The theoretical results suggest that this function approximates well classic rigorous solutions. For such loss function, we propose an end-to-end method for the training of deep representation learning CPD models. Our experiments provide evidence that the proposed approach improves baseline results of change point detection for various data types, including real-world videos and image sequences, and improve representations for them. * This work is supported by the Russian Science Foundation (project 20-71-10135) Preprint. Under review.

Usage of specific attention improves change point detection

“…Also, we apply different masks for the source sequence. To compare with [10] we use the RNN model as a baseline. Intuitively, it seems that if we have only one point of change, we do not need to look at the entire sequence of data, but we can only consider some part of it.…”

“…The solution of this problem with the neural network was investigated in [10]. Its working principle is as follows.…”

“…According to [10] the equation ( 3) is the lower bound for the expected value of detection delay and the equation ( 2) is lower bound for the expected time to false alarm.…”

“…Such tasks appear in different areas, from monitoring systems to video analysis [1] to oil&gas [11]. One of the recent methods for CPD [10] proves that a recurrent neural network model constructs meaningful representations and solve a problem better than non-principled approaches. The state-of-the-art model for sequential data is Transformer [17].…”

“…Simple statistics-based approaches are sufficient in many cases [4]. However, semi-structured data require deep model to provide reasonable quality of solutions ranging from rather simple neural network [6] to more complex workflows [10,7] and problem statements [14].…”