SPROCKET: Extending ROCKET to Distance-Based Time-Series Transformations With Prototypes
By: Nicholas Harner
Potential Business Impact:
Finds patterns in data faster and better.
Classical Time Series Classification algorithms are dominated by feature engineering strategies. One of the most prominent of these transforms is ROCKET, which achieves strong performance through random kernel features. We introduce SPROCKET (Selected Prototype Random Convolutional Kernel Transform), which implements a new feature engineering strategy based on prototypes. On a majority of the UCR and UEA Time Series Classification archives, SPROCKET achieves performance comparable to existing convolutional algorithms and the new MR-HY-SP ( MultiROCKET-HYDRA-SPROCKET) ensemble's average accuracy ranking exceeds HYDRA-MR, the previous best convolutional ensemble's performance. These experimental results demonstrate that prototype-based feature transformation can enhance both accuracy and robustness in time series classification.
Similar Papers
Signature Kernel Scoring Rule as Spatio-Temporal Diagnostic for Probabilistic Forecasting
Machine Learning (Stat)
Improves weather forecasts by understanding how weather changes.
ProtoTSNet: Interpretable Multivariate Time Series Classification With Prototypical Parts
Machine Learning (CS)
Helps doctors understand patient health data.
Accelerating Time Series Foundation Models with Speculative Decoding
Machine Learning (CS)
Speeds up predictions for websites and apps.