公开(公告)号：US20240320529A1

公开(公告)日：2024-09-26

申请号：US18611417

申请日：2024-03-20

Applicant: DeepMind Technologies Limited

Inventor： Sumanth Dathathri ,  Abigail Elizabeth See ,  Borja De Balle Pigem ,  Sumedh Kedar Ghaisas ,  Pushmeet Kohli ,  Po-Sen Huang ,  Johannes Maximilian Welbl

IPC: G06N7/01

CPC classification number: G06N7/01

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for watermarking a digital object generated by a machine learning model. The digital object is defined by a sequence of tokens. The watermarking involves modifying a probability distribution of the tokens by applying a succession of watermarking stages.

公开(公告)号：US11636347B2

公开(公告)日：2023-04-25

申请号：US16749252

申请日：2020-01-22

Applicant: DeepMind Technologies Limited

Inventor： Hanjun Dai ,  Yujia Li ,  Chenglong Wang ,  Rishabh Singh ,  Po-Sen Huang ,  Pushmeet Kohli

IPC: G06N3/08 ,  G06K9/62 ,  G06N3/04 ,  G06N3/088 ,  G06N3/049

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting actions to be performed by an agent interacting with an environment. In one aspect, a method comprises: obtaining a graph of nodes and edges that represents an interaction history of the agent with the environment; generating an encoded representation of the graph representing the interaction history of the agent with the environment; processing an input based on the encoded representation of the graph using an action selection neural network, in accordance with current values of action selection neural network parameters, to generate an action selection output; and selecting an action from a plurality of possible actions to be performed by the agent using the action selection output generated by the action selection neural network.

公开(公告)号：US20200234145A1

公开(公告)日：2020-07-23

申请号：US16749252

申请日：2020-01-22

Applicant: DeepMind Technologies Limited

Inventor： Hanjun Dai ,  Yujia Li ,  Chenglong Wang ,  Rishabh Singh ,  Po-Sen Huang ,  Pushmeet Kohli

IPC: G06N3/08 ,  G06N3/04 ,  G06K9/62

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting actions to be performed by an agent interacting with an environment. In one aspect, a method comprises: obtaining a graph of nodes and edges that represents an interaction history of the agent with the environment; generating an encoded representation of the graph representing the interaction history of the agent with the environment; processing an input based on the encoded representation of the graph using an action selection neural network, in accordance with current values of action selection neural network parameters, to generate an action selection output; and selecting an action from a plurality of possible actions to be performed by the agent using the action selection output generated by the action selection neural network.