Gaussian processes based data augmentation and expected signature for time series classification
This work addresses time series classification, likely for domains like finance or healthcare, but appears incremental as it builds on existing signature and Gaussian processes methods.
The authors tackled time series classification by proposing a feature extraction model based on the expected signature, computed via Gaussian processes-based data augmentation, with the model learning optimal features through supervised tasks.
The signature is a fundamental object that describes paths (that is, continuous functions from an interval to a Euclidean space). Likewise, the expected signature provides a statistical description of the law of stochastic processes. We propose a feature extraction model for time series built upon the expected signature. This is computed through a Gaussian processes based data augmentation. One of the main features is that an optimal feature extraction is learnt through the supervised task that uses the model.