HIT-ROCKET: Hadamard-vector Inner-product Transformer for ROCKET
This work addresses efficiency and performance issues in time series classification for applications like communications and finance, representing an incremental improvement over existing lightweight methods.
The paper tackles the high computational complexity and training time of state-of-the-art time series classification methods by proposing a Hadamard-vector inner-product transformer, which improves F1-score by at least 5% over ROCKET and reduces training time by 50% compared to miniROCKET.
Time series classification holds broad application value in communications, information countermeasures, finance, and medicine. However, state-of-the-art (SOTA) methods-including HIVE-COTE, Proximity Forest, and TS-CHIEF-exhibit high computational complexity, coupled with lengthy parameter tuning and training cycles. In contrast, lightweight solutions like ROCKET (Random Convolutional Kernel Transform) offer greater efficiency but leave substantial room for improvement in kernel selection and computational overhead. To address these challenges, we propose a feature extraction approach based on Hadamard convolutional transform, utilizing column or row vectors of Hadamard matrices as convolution kernels with extended lengths of varying sizes. This enhancement maintains full compatibility with existing methods (e.g., ROCKET) while leveraging kernel orthogonality to boost computational efficiency, robustness, and adaptability. Comprehensive experiments on multi-domain datasets-focusing on the UCR time series dataset-demonstrate SOTA performance: F1-score improved by at least 5% vs. ROCKET, with 50% shorter training time than miniROCKET (fastest ROCKET variant) under identical hyperparameters, enabling deployment on ultra-low-power embedded devices. All code is available on GitHub.