Generalised Interpretable Shapelets for Irregular Time Series
This work addresses the problem of time series analysis for researchers and practitioners by providing a more flexible and interpretable method, though it appears incremental as it builds on existing shapelet techniques.
The authors tackled the limitations of shapelet transforms for time series by extending them to continuous time, enabling handling of irregularly-sampled, partially-observed multivariate data, and introduced a regularization penalty for efficient training without losing interpretability, achieving performance validated on datasets like digit and audio classification.
The shapelet transform is a form of feature extraction for time series, in which a time series is described by its similarity to each of a collection of `shapelets'. However it has previously suffered from a number of limitations, such as being limited to regularly-spaced fully-observed time series, and having to choose between efficient training and interpretability. Here, we extend the method to continuous time, and in doing so handle the general case of irregularly-sampled partially-observed multivariate time series. Furthermore, we show that a simple regularisation penalty may be used to train efficiently without sacrificing interpretability. The continuous-time formulation additionally allows for learning the length of each shapelet (previously a discrete object) in a differentiable manner. Finally, we demonstrate that the measure of similarity between time series may be generalised to a learnt pseudometric. We validate our method by demonstrating its performance and interpretability on several datasets; for example we discover (purely from data) that the digits 5 and 6 may be distinguished by the chirality of their bottom loop, and that a kind of spectral gap exists in spoken audio classification.