Enabling Tensor Decomposition for Time-Series Classification via A Simple Pseudo-Laplacian Contrast

TL;DR

This paper introduces a Pseudo Laplacian Contrast tensor decomposition method that enhances time-series classification by extracting class-aware features through graph-based regularization, improving low-rank representation learning.

Contribution

It proposes a novel PLC tensor decomposition framework that integrates data augmentation and Laplacian regularization for improved classification, addressing tensor non-uniqueness issues.

Findings

Effective class-aware feature extraction demonstrated on multiple datasets

Outperforms existing tensor decomposition methods in classification tasks

Unsupervised optimization algorithm efficiently estimates pseudo graphs

Abstract

Tensor decomposition has emerged as a prominent technique to learn low-dimensional representation under the supervision of reconstruction error, primarily benefiting data inference tasks like completion and imputation, but not classification task. We argue that the non-uniqueness and rotation invariance of tensor decomposition allow us to identify the directions with largest class-variability and simple graph Laplacian can effectively achieve this objective. Therefore we propose a novel Pseudo Laplacian Contrast (PLC) tensor decomposition framework, which integrates the data augmentation and cross-view Laplacian to enable the extraction of class-aware representations while effectively capturing the intrinsic low-rank structure within reconstruction constraint. An unsupervised alternative optimization algorithm is further developed to iteratively estimate the pseudo graph and minimize the loss using Alternating Least Square (ALS). Extensive experimental results on various datasets demonstrate the effectiveness of our approach.