A GA-based feature selection and parameter optimization for linear support higher-order tensor machine

• We propose a GA-based feature selection and parameter optimization algorithm for SHTM.
• The proposed GA is coded according to the tensor rank-one decomposition.
• The testing accuracy of the proposed algorithm is significantly higher than that of the linear SHTM.
• The training time of the proposed algorithm is usually shorter than that of the linear SHTM for higher-order tensors.
• The proposed algorithm can remove the redundancy information of tensor data and improve the testing accuracy of SHTM.

In the fields of pattern recognition, computer vision, and image processing, many real-world image and video data are more naturally represented as tensors. Recently, based on the supervised tensor learning (STL) framework, a linear support higher-order tensor machine (SHTM) has been proposed. Considering that there are much redundancy information in the tensor data and the model parameter largely affects the performance of SHTM, in this study, we present a genetic algorithm (GA) based feature selection and parameter optimization algorithm for the linear SHTM. The proposed algorithm can remove the redundancy information in tensor data and obtain a better generalized accuracy by searching for the optimal model parameter and feature subset simultaneously. A set of experiments is conducted on nine second-order face recognition datasets and three third-order gait recognition datasets to illustrate the performance of the proposed algorithm. The statistic test shows that compared with the original linear SHTM, the proposed algorithm can provide a significant performance gain in terms of generalized accuracy for tensor classification.