Microsoft Technology Licensing, LLC

Baidu USA LLC

Venmugil Elango

Douglas Christopher Burger

Eric S Chung

Bita Darvish Rouhani

Ming Gang Liu

Jinwen Xi

Maximilian Golub

Jeremy Fowers

Rasoul Shafipour

Embodiments of the present disclosure include systems and methods for performing data-aware model pruning for neural networks. During a training phase, a neural network is trained with a first set of data. During a validation phase, inference with the neural network is performed using a second set of data that causes the neural network to generate a first set of outputs at a layer in the neural network. During the validation phase, a plurality of mean values and a plurality of variance values are calculated based on the first set of outputs. A plurality of entropy values are calculated based on the plurality of mean values and the plurality of variance values. A second set of outputs are pruned based on the plurality of entropy values. The second set of outputs are generated by the layer of the neural network using a third set of data.

Data-aware model pruning for neural networks

Presented are systems and methods to automatically find efficient parallelization strategies for deep neural networks (DNNs). A computation graph comprising an efficiently ordered sequence of vertices aids in computing the best parallelizing strategy in a relatively short time. Effectiveness of the parallelization strategies is evaluated on various DNNs, and the performance of the strategies proposed by various embodiments is compared against data parallelism, expert-designed strategies, and other state-of-the-art approaches. Experimental results demonstrate that the proposed strategies outperform a baseline data parallelism strategy and achieve better performance than expert-designed strategies and state-of-the-art approaches.

Reducing training times of deep neural networks through efficient hybrid parallelism

Embodiments of the present disclosure include systems and methods for providing hierarchical and shared exponent floating point data types. First and second shared exponent values are determined based on exponent values of a plurality of floating point values. A third shared exponent value is determined based the first shared exponent value and the second shared exponent value. First and second difference values are determined based on the first shared exponent value, the second shared exponent value, and the third shared exponent value. Sign values and mantissa values are determined for the plurality of floating point values. The sign value and the mantissa value for each floating point value in the plurality of floating point values, the third shared exponent value, the first difference value, and the second difference value are stored in a data structure for a shared exponent floating point data type.

Hierarchical and shared exponent floating point data types

Growing community of inventors

Redmond, WA, United States of America

Venmugil Elango