Abstract:Canonical polyadic decomposition (CPD) is one of the most common tensor computations adopted in many scientific applications. The major bottleneck of CPD is matricized tensor times Khatri-Rao product (MTTKRP). To optimize the performance of MTTKRP, various sparse tensor formats have been proposed such as CSF and HiCOO. However, due to the spatial complexity of the tensors, no single format fits all tensors. To address this problem, we propose SpTFS, a framework that automatically predicts the optimal storage format for an input sparse tensor. Specifically, SpTFS leverages a set of sampling methods to lower the sparse tensor to fixed-sized matrices and sparsity features. In addition, SpTFS adopts both supervised learning based and unsupervised learning based methods to predict the optimal sparse tensor storage formats. For supervised learning, we propose TnsNet that combines convolution neural network (CNN) and the feature layer, which effectively captures the sparsity patterns of the input tensors. Once trained, the TnsNet can be used with either density or histogram representation of the input tensor for optimal format prediction. Whereas for unsupervised learning, we propose TnsClustering that consists of a feature encoder using convolutional layers and fully connected layers, and a K-means++ model to cluster sparse tensors for optimal tensor format prediction, without massively profiling on the hardware platform. SpTFS can use the above two models to predict the optimal tensor storage format for accelerating MTTKRP accurately. The experimental results show that both TnsNet and TnsClustering can achieve higher prediction accuracy and performance speedup compared to the state-of-the-art works.

Tensor LISTA: Differentiable sparse representation learning for multi-dimensional tensor

Compressed Lista Exploiting Toeplitz Structure

Efficient Multi-Dimensional Tensor Sparse Coding Using t-Linear Combination.

Tensor low-rank sparse representation for tensor subspace learning

Learned Extragradient ISTA with Interpretable Residual Structures for Sparse Coding.

Spatial-Temporal Tensor Representation Learning with Priors for Infrared Small Target Detection

Learning Separable Dictionaries for Sparse Tensor Representation: an Online Approach

Learning with Tensor Representation

Sparse tensor discriminant analysis

Tensor Decomposition for Signal Processing and Machine Learning

Tensor Low-Rank Representation for Data Recovery and Clustering

Structured LISTA for Multidimensional Harmonic Retrieval

Tensor-based Intrinsic Subspace Representation Learning for Multi-view Clustering

SpTFS: Sparse Tensor Format Selection for MTTKRP Via Deep Learning

Variational Learning ISTA

Input-aware Sparse Tensor Storage Format Selection for Optimizing MTTKRP

Learning tensor networks with tensor cross interpolation: new algorithms and libraries

Optimal Calculation of Tensor Learning Approaches.

Optimum Subspace Learning And Error Correction For Tensors

Low-Rank Tensor Train Coefficient Array Estimation for Tensor-on-Tensor Regression

An algorithm for online tensor prediction