LOW DISPLACEMENT RANK BASED DEEP NEURAL NETWORK COMPRESSION

摘要

A method and an apparatus for performing deep neural network compression use an approximation training set along with information, such as in matrices representing weights, biases and non-linearities, to iteratively compress a p re-trained deep neural network by low displacement rank based approximation of the network layer weight matrices. The low displacement rank approximation allows for replacement of an original layer weight matrices of the pre-trained deep neural network as the sum of a small number of structured matrices, allowing compression and low inference complexity.