ATD: Augmenting CP Tensor Decomposition by Self Supervision
This work addresses the problem of improving classification performance in tensor-based dimensionality reduction for researchers and practitioners in signal processing and machine learning, representing an incremental advancement by combining existing techniques.
The paper tackles the misalignment between traditional tensor decomposition objectives and downstream classification tasks by proposing Augmented Tensor Decomposition (ATD), which integrates data augmentations and self-supervised learning, resulting in accuracy gains of 0.8% to 2.5% over tensor baselines and up to 15% over other models with fewer parameters.
Tensor decompositions are powerful tools for dimensionality reduction and feature interpretation of multidimensional data such as signals. Existing tensor decomposition objectives (e.g., Frobenius norm) are designed for fitting raw data under statistical assumptions, which may not align with downstream classification tasks. In practice, raw input tensors can contain irrelevant information while data augmentation techniques may be used to smooth out class-irrelevant noise in samples. This paper addresses the above challenges by proposing augmented tensor decomposition (ATD), which effectively incorporates data augmentations and self-supervised learning (SSL) to boost downstream classification. To address the non-convexity of the new augmented objective, we develop an iterative method that enables the optimization to follow an alternating least squares (ALS) fashion. We evaluate our proposed ATD on multiple datasets. It can achieve 0.8% - 2.5% accuracy gain over tensor-based baselines. Also, our ATD model shows comparable or better performance (e.g., up to 15% in accuracy) over self-supervised and autoencoder baselines while using less than 5% of learnable parameters of these baseline models