Multiscale Analysis for Higher-order Tensors
For researchers working with high-order tensor data, this method addresses the limitation of existing linear low-rank approximations by enabling adaptive, multi-scale decomposition, though the improvements are demonstrated only on specific datasets without quantitative comparison to state-of-the-art.
The paper introduces a multi-scale higher-order singular value decomposition (MS-HoSVD) method for tensor data, which adaptively partitions tensors into low-rank sub-tensors to improve representational efficiency. The method shows promising results in dimensionality reduction and classification tasks on real-life tensor signals.
The widespread use of multisensor technology and the emergence of big datasets have created the need to develop tools to reduce, approximate, and classify large and multimodal data such as higher-order tensors. While early approaches focused on matrix and vector based methods to represent these higher-order data, more recently it has been shown that tensor decomposition methods are better equipped to capture couplings across their different modes. For these reasons, tensor decomposition methods have found applications in many different signal processing problems including dimensionality reduction, signal separation, linear regression, feature extraction, and classification. However, most of the existing tensor decomposition methods are based on the principle of finding a low-rank approximation in a linear subspace structure, where the definition of the rank may change depending on the particular decomposition. Since many datasets are not necessarily low-rank in a linear subspace, this often results in high approximation errors or low compression rates. In this paper, we introduce a new adaptive, multi-scale tensor decomposition method for higher order data inspired by hybrid linear modeling and subspace clustering techniques. In particular, we develop a multi-scale higher-order singular value decomposition (MS-HoSVD) approach where a given tensor is first permuted and then partitioned into several sub-tensors each of which can be represented as a low-rank tensor with increased representational efficiency. The proposed approach is evaluated for dimensionality reduction and classification for several different real-life tensor signals with promising results.