公开(公告)号：US10748039B2

公开(公告)日：2020-08-18

申请号：US16586262

申请日：2019-09-27

Applicant: DeepMind Technologies Limited

Inventor： Catalin-Dumitru Ionescu ,  Tejas Dattatraya Kulkarni

IPC: G06K9/62 ,  G06F16/56 ,  G06N3/04 ,  G06N3/08

Abstract: A reinforcement learning neural network system in which internal representations and policies are grounded in visual entities derived from image pixels comprises a visual entity identifying neural network subsystem configured to process image data to determine a set of spatial maps representing respective discrete visual entities. A reinforcement learning neural network subsystem processes data from the set of spatial maps and environmental reward data to provide action data for selecting actions to perform a task.

公开(公告)号：US20220215580A1

公开(公告)日：2022-07-07

申请号：US17608620

申请日：2020-05-05

Applicant: DeepMind Technologies Limited

Inventor： Ankush Gupta ,  Tejas Dattatraya Kulkarni

IPC: G06T7/73

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for unsupervised learning of object keypoint locations in images. In particular, a keypoint extraction machine learning model having a plurality of keypoint model parameters is trained to receive an input image and to process the input image in accordance with the keypoint model parameters to generate a plurality of keypoint locations in the input image. The machine learning model is trained using either temporal transport or spatio-temporal transport.

公开(公告)号：US20200104645A1

公开(公告)日：2020-04-02

申请号：US16586262

申请日：2019-09-27

Applicant: DeepMind Technologies Limited

Inventor： Catalin-Dumitru Ionescu ,  Tejas Dattatraya Kulkarni

IPC: G06K9/62 ,  G06F16/56 ,  G06N3/08 ,  G06N3/04

Abstract: A reinforcement learning neural network system in which internal representations and policies are grounded in visual entities derived from image pixels comprises a visual entity identifying neural network subsystem configured to process image data to determine a set of spatial maps representing respective discrete visual entities. A reinforcement learning neural network subsystem processes data from the set of spatial maps and environmental reward data to provide action data for selecting actions to perform a task.

公开(公告)号：US12175723B2

公开(公告)日：2024-12-24

申请号：US17608620

申请日：2020-05-05

Applicant: DeepMind Technologies Limited

Inventor： Ankush Gupta ,  Tejas Dattatraya Kulkarni

IPC: G06V10/00 ,  G06T7/73 ,  G06V10/46 ,  G06V10/82

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for unsupervised learning of object keypoint locations in images. In particular, a keypoint extraction machine learning model having a plurality of keypoint model parameters is trained to receive an input image and to process the input image in accordance with the keypoint model parameters to generate a plurality of keypoint locations in the input image. The machine learning model is trained using either temporal transport or spatio-temporal transport.

公开(公告)号：US20210271968A1

公开(公告)日：2021-09-02

申请号：US16967597

申请日：2019-02-11

Applicant: DeepMind Technologies Limited

Inventor： Iaroslav Ganin ,  Tejas Dattatraya Kulkarni ,  Oriol Vinyals ,  Seyed Mohammadali Eslami

IPC: G06N3/08 ,  G06K9/62 ,  G06F30/20

Abstract: A generative adversarial neural network system to provide a sequence of actions for performing a task. The system comprises a reinforcement learning neural network subsystem coupled to a simulator and a discriminator neural network. The reinforcement learning neural network subsystem includes a policy recurrent neural network to, at each of a sequence of time steps, select one or more actions to be performed according to an action selection policy, each action comprising one or more control commands for a simulator. The simulator is configured to implement the control commands for the time steps to generate a simulator output. The discriminator neural network is configured to discriminate between the simulator output and training data, to provide a reward signal for the reinforcement learning. The simulator may be non-differentiable simulator, for example a computer program to produce an image or audio waveform or a program to control a robot or vehicle.