公开(公告)号：US20220129759A1

公开(公告)日：2022-04-28

申请号：US17441622

申请日：2019-06-26

Applicant: Intel Corporation

Inventor： Anbang YAO ,  Aojun ZHOU ,  Dawei SUN ,  Dian GU ,  Yurong CHEN

IPC: G06N3/08 ,  G06T9/00 ,  G06N3/04

Abstract: Apparatuses, methods, and GPUs are disclosed for universal loss-error-aware quantization (ULQ) of a neural network (NN). In one example, an apparatus includes data storage to store data including activation sets and weight sets, and a network processor coupled to the data storage. The network processor is configured to implement the ULQ by constraining a low-precision NN model based on a full-precision NN model, to perform a loss-error-aware activation quantization to quantize activation sets into ultra-low-bit versions with given bit-width values, to optimize the NN with respect to a loss function that is based on the full-precision NN model, and to perform a loss-error-aware weight quantization to quantize weight sets into ultra-low-bit versions.

公开(公告)号：US20220147791A1

公开(公告)日：2022-05-12

申请号：US17435657

申请日：2019-06-21

Applicant: Intel Corporation

Inventor： Anbang YAO ,  Jiahui ZHANG ,  Dawei SUN ,  Dian GU ,  Yurong CHEN

IPC: G06N3/04

Abstract: Embodiments are generally directed to sparse 3D convolution acceleration in a convolutional layer of an artificial neural network model. An embodiment of an apparatus includes one or more processors including a graphics processor to process data; and a memory for storage of data, including feature maps. The one or more processors are to provide for sparse 3D convolution acceleration by applying a shared 3D convolutional kernel/filter to an input feature map to produce an output feature map, including increasing sparsity of the input feature map by partitioning it into multiple disjoint input groups; generation of multiple disjoint output groups corresponding to the input groups by performing a convolution calculation represented by the shared 3D convolutional kernel/filter on all feature values associated with active/valid voxels of each input group to produce corresponding feature values within corresponding output groups; and outputting the output feature map by sequentially stacking the output groups.