公开(公告)号：US20230359865A1

公开(公告)日：2023-11-09

申请号：US18044842

申请日：2020-09-16

Applicant: Google LLC

Inventor： Zhuoran Shen ,  Raviteja Vemulapalli ,  Irwan Bello ,  Xuhui Jia ,  Ching-Hui Chen

IPC: G06N3/045 ,  G06N3/084

CPC classification number: G06N3/045 ,  G06N3/084

Abstract: The present disclosure provides systems, methods, and computer program products for modeling dependencies throughout a network using a global-self attention model with a content attention layer and a positional attention layer that operate in parallel. The model receives input data comprising content values and context positions. The content attention layer generates one or more output features for each context position based on a global attention operation applied to the content values independent of the context positions. The positional attention layer generates an attention map for each of the context positions based on one or more content values of the respective context position and associated neighboring positions. Output is determined based on the output features generated by the content attention layer and the attention map generated for each context position by the positional attention layer. The model improves efficiency and can be used throughout a deep network.

公开(公告)号：US20230281979A1

公开(公告)日：2023-09-07

申请号：US18006078

申请日：2020-08-03

Applicant: Xuhui JIA ,  Raviteja VEMULAPALLI ,  Yukun ZHU ,  Bradley Ray GREEN ,  Bardia DOOSTI ,  Ching-Hui CHEN ,  Google LLC

Inventor： Xuhui Jia ,  Raviteja Vemulapalli ,  Bradley Ray Green ,  Bardia Doosti ,  Ching-Hui Chen

IPC: G06V10/82 ,  G06V10/776

CPC classification number: G06V10/82 ,  G06V10/776

Abstract: Systems and methods of the present disclosure are directed to a method for training a machine-learned visual attention model. The method can include obtaining image data that depicts a head of a person and an additional entity. The method can include processing the image data with an encoder portion of the visual attention model to obtain latent head and entity encodings. The method can include processing the latent encodings with the visual attention model to obtain a visual attention value and processing the latent encodings with a machine-learned visual location model to obtain a visual location estimation. The method can include training the models by evaluating a loss function that evaluates differences between the visual location estimation and a pseudo visual location label derived from the image data and between the visual attention value and a ground truth visual attention label.

公开(公告)号：US20230214656A1

公开(公告)日：2023-07-06

申请号：US18009629

申请日：2020-06-10

Applicant: Google LLC

Inventor： Raviteja Vemulapalli ,  Jianrui Cai ,  Bradley Ray Green ,  Ching-Hui Chen ,  Lior Shapira

IPC: G06N3/082 ,  G06V10/82 ,  G06V10/764

CPC classification number: G06N3/082 ,  G06V10/82 ,  G06V10/764

Abstract: At training time, a base neural network can be trained to perform each of a plurality of basis subtasks included in a total set of basis subtasks (e.g., individually or some combination thereof). Next, a description of a desired combined subtask can be obtained. Based on the description of the combined subtask, a mask generator can produce a pruning mask which is used to prune the base neural network into a smaller combined-subtask-specific network that performs only the two or more basis subtasks included in the combined subtask.