Common Mode Patterns for Supervised Tensor Subspace Learning
This work addresses a domain-specific problem in hyperspectral imaging by providing an incremental improvement in supervised tensor subspace learning for classification tasks.
The paper tackled the problem of dimensionality reduction for tensor objects in binary classification by proposing the Common Mode Patterns method, which uses label information to prevent different classes from sharing common features after reduction, and experimentally showed it efficiently reduces dimensionality while increasing inter-class separability compared to Multilinear Principal Component Analysis on a hyperspectral imaging dataset.
In this work we propose a method for reducing the dimensionality of tensor objects in a binary classification framework. The proposed Common Mode Patterns method takes into consideration the labels' information, and ensures that tensor objects that belong to different classes do not share common features after the reduction of their dimensionality. We experimentally validate the proposed supervised subspace learning technique and compared it against Multilinear Principal Component Analysis using a publicly available hyperspectral imaging dataset. Experimental results indicate that the proposed CMP method can efficiently reduce the dimensionality of tensor objects, while, at the same time, increasing the inter-class separability.