公开(公告)号：US11816185B1

公开(公告)日：2023-11-14

申请号：US16383347

申请日：2019-04-12

Applicant: Nvidia Corporation

Inventor： Holger Roth ,  Yingda Xia ,  Dong Yang ,  Daguang Xu

IPC: G06F18/214 ,  G06F9/30 ,  G06N3/08 ,  G16H30/40 ,  G06N5/04 ,  G06F18/211 ,  G06F18/2433 ,  G06N3/045

CPC classification number: G06F18/2155 ,  G06F9/3001 ,  G06F18/211 ,  G06F18/2433 ,  G06N3/045 ,  G06N3/08 ,  G06N5/04 ,  G16H30/40 ,  G06V2201/031

Abstract: Volumetric quantification can be performed for various parameters of an object represented in volumetric data. Multiple views of the object can be generated, and those views provided to a set of neural networks that can generate inferences in parallel. The inferences from the different networks can be used to generate pseudo-labels for the data, for comparison purposes, which enables a co-training loss to be determined for the unlabeled data. The co-training loss can then be used to update the relevant network parameters for the overall data analysis network. If supervised data is also available then the network parameters can further be updated using the supervised loss.

公开(公告)号：US20200327674A1

公开(公告)日：2020-10-15

申请号：US16380759

申请日：2019-04-10

Applicant: NVIDIA Corporation

Inventor： Dong Yang ,  Daguang Xu ,  Fengze Liu ,  Yingda Xia

IPC: G06T7/12 ,  G06N3/08 ,  G06T7/13 ,  G06T7/60 ,  G06K9/62

Abstract: Comparison logic compares boundaries of features of or more images based, at least in part, on identifying boundaries and indication logic coupled to the comparison logic to indicate whether the boundaries differ by at least a first threshold. The boundaries might comprise a first label mask representing boundaries of objects in an image that are boundaries in a segmentation determined from a segmentation process and a second label mask from a shape evaluation process applied to the first label mask. The indication logic might be configured to compare the first label mask and the second label mask to determine a quality of the segmentation. A neural network might perform the segmentation. Shape evaluation using the first label mask as an input and the second label mask as an output might be performed by a variational autoencoder. A graphical processing unit (GPU) might be used for the segmentation and/or the autoencoder.

公开(公告)号：US12164599B1

公开(公告)日：2024-12-10

申请号：US18232202

申请日：2023-08-09

Applicant: NVIDIA Corporation

Inventor： Holger Roth ,  Yingda Xia ,  Dong Yang ,  Daguang Xu

IPC: G06F18/214 ,  G06F9/30 ,  G06F18/211 ,  G06F18/2433 ,  G06N3/045 ,  G06N3/08 ,  G06N5/04 ,  G16H30/40

Abstract: Volumetric quantification can be performed for various parameters of an object represented in volumetric data. Multiple views of the object can be generated, and those views provided to a set of neural networks that can generate inferences in parallel. The inferences from the different networks can be used to generate pseudo-labels for the data, for comparison purposes, which enables a co-training loss to be determined for the unlabeled data. The co-training loss can then be used to update the relevant network parameters for the overall data analysis network. If supervised data is also available then the network parameters can further be updated using the supervised loss.

公开(公告)号：US20220366220A1

公开(公告)日：2022-11-17

申请号：US17244781

申请日：2021-04-29

Applicant: NVIDIA Corporation

Inventor： Holger Reinhard Roth ,  Yingda Xia ,  Daguang Xu ,  Andriy Myronenko ,  Wenqi Li ,  Dong Yang

IPC: G06N3/04 ,  H04L29/08 ,  G06K9/62

Abstract: Apparatuses, systems, and techniques to improve federated learning for neural networks. In at least one embodiment, a federated server dynamically selects neural network weights according to one or more learnable aggregation weights indicating a contribution from each of one or more edge devices or clients during federated training according to various characteristics of each edge device or client model and training data.