Sparse Deep Neural Network Exact Solutions
arXiv:1807.03165 · doi:10.1109/HPEC.2018.8547742
Abstract
Deep neural networks (DNNs) have emerged as key enablers of machine learning. Applying larger DNNs to more diverse applications is an important challenge. The computations performed during DNN training and inference are dominated by operations on the weight matrices describing the DNN. As DNNs incorporate more layers and more neurons per layers, these weight matrices may be required to be sparse because of memory limitations. Sparse DNNs are one possible approach, but the underlying theory is in the early stages of development and presents a number of challenges, including determining the accuracy of inference and selecting nonzero weights for training. Associative array algebra has been developed by the big data community to combine and extend database, matrix, and graph/network concepts for use in large, sparse data problems. Applying this mathematics to DNNs simplifies the formulation of DNN mathematics and reveals that DNNs are linear over oscillating semirings. This work uses associative array DNNs to construct exact solutions and corresponding perturbation models to the rectified linear unit (ReLU) DNN equations that can be used to construct test vectors for sparse DNN implementations over various precisions. These solutions can be used for DNN verification, theoretical explorations of DNN properties, and a starting point for the challenge of sparse training.
arXiv comment8 pages, 10 figures, accepted to IEEE HPEC 2018. arXiv admin note: text overlap with arXiv:1708.02937
References in corpus (6)
- SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
- Exact solutions to the nonlinear dynamics of learning in deep linear neural networks
- Theory II: Landscape of the Empirical Risk in Deep Learning
- Enabling Massive Deep Neural Networks with the GraphBLAS
- Speeding up Convolutional Neural Networks By Exploiting the Sparsity of Rectifier Units
- Polystore Mathematics of Relational Algebra