Universal hidden monotonic trend estimation with contrastive learning
This provides a flexible tool for trend analysis across various domains, though it appears incremental as it builds upon the Mann-Kendall test.
The paper tackles the problem of extracting hidden monotonic trends from diverse temporal data types without standard assumptions, proposing Contrastive Trend Estimation (CTE) as a universal method that works on vectors, images, graphs, and time series.
In this paper, we describe a universal method for extracting the underlying monotonic trend factor from time series data. We propose an approach related to the Mann-Kendall test, a standard monotonic trend detection method and call it contrastive trend estimation (CTE). We show that the CTE method identifies any hidden trend underlying temporal data while avoiding the standard assumptions used for monotonic trend identification. In particular, CTE can take any type of temporal data (vector, images, graphs, time series, etc.) as input. We finally illustrate the interest of our CTE method through several experiments on different types of data and problems.