HYDRA: Competing convolutional kernels for fast and accurate time series classification
This work addresses time series classification for researchers and practitioners, offering an incremental improvement by bridging and enhancing existing methods.
The paper tackled the problem of time series classification by connecting dictionary methods and convolutional kernel methods like ROCKET, introducing HYDRA as a fast and accurate dictionary method that combines aspects of both. The result shows HYDRA is faster and more accurate than existing dictionary methods, with potential for further accuracy improvements when combined with ROCKET variants.
We demonstrate a simple connection between dictionary methods for time series classification, which involve extracting and counting symbolic patterns in time series, and methods based on transforming input time series using convolutional kernels, namely ROCKET and its variants. We show that by adjusting a single hyperparameter it is possible to move by degrees between models resembling dictionary methods and models resembling ROCKET. We present HYDRA, a simple, fast, and accurate dictionary method for time series classification using competing convolutional kernels, combining key aspects of both ROCKET and conventional dictionary methods. HYDRA is faster and more accurate than the most accurate existing dictionary methods, and can be combined with ROCKET and its variants to further improve the accuracy of these methods.