Alibaba Group Holding Limited

Zhenyu Gu

Yuan Xie

Shuangchen Li

Tao Zhang

Qiong Zhang

Luo Si

Xiaogang Li

Maohua Zhu

Liu Liu

Huasha Zhao

Qiyu Zhang

The present disclosure relates to systems and methods for providing vector-wise sparsity in neural networks. In some embodiments, an exemplary method for providing vector-wise sparsity in a neural network, comprises: dividing a matrix associated with the neural network into a plurality of vectors; selecting a first subset of non-zero elements from the plurality of vectors to form a pruned matrix; and outputting the pruned matrix for executing the neural network using the pruned matrix.

Systems and methods for providing vector-wise sparsity in a neural network

Systems and methods for dynamically executing sparse neural networks are provided. In one implementation, a system for providing dynamic sparsity in a neural network may include at least one memory storing instructions and at least one processor configured to execute the instructions to: reduce an input vector and a set of weights of the neural network, execute an input layer of the neural network using the reduced input vector and set of weights to generate a reduced output vector; expand the reduced output vector to a full output vector using first predictable output neurons (PONs); using a PON map, reduce a dimension of the full output vector; execute subsequent layers of the neural network using the reduced full output vector to produce a second reduced output vector; and expand the second reduced output vector to a second full output vector using second PONs.

Systems and methods for accelerating sparse neural network execution

The present disclosure relates to systems and methods for providing block-wise sparsity in neural networks. In one implementation, a system for providing block-wise sparsity in a neural network may include at least one memory storing instructions and at least one processor configured to execute the instructions to: divide a matrix of weights associated with a neural network into a plurality of blocks; extract non-zero elements from one or more of the plurality of blocks; re-encode the extracted non-zero elements as vectors with associated coordinates of the extracted non-zero elements within the one or more blocks; enforce input sparsity in the neural network corresponding to the associated coordinates; and execute the neural network using the vectors and the enforced input sparsity.

Systems and methods for providing block-wise sparsity in a neural network

A method for generating one or more push notifications to a user device is described. The method comprises: obtaining history data representing a history of online activities of a user and candidate data representing a set of candidate information; generating, based on the history data and the candidate data, user profile vectors representing a user profile associated with the user and content vectors representing a set of content profiles associated with the set of candidate information; generating, based on a machine learning model trained with a history of online activities, embedding user feature vectors and embedding content feature vectors based on the history data and the candidate data; and providing for transmission information for one or more push notifications including first candidate information of to a user device associated with the user, the first candidate information being determined from the set of candidate information based on the aforementioned vectors.

Method and apparatus for generating push notifications

Growing community of inventors

San Mateo, CA, United States of America

Zhenyu Gu