公开(公告)号：US20240281654A1

公开(公告)日：2024-08-22

申请号：US18292165

申请日：2022-08-12

Applicant: DeepMind Technologies Limited

Inventor： Scott Ellison Reed ,  Konrad Zolna ,  Emilio Parisotto ,  Tom Erez ,  Alexander Novikov ,  Jack William Rae ,  Misha Man Ray Denil ,  Joao Ferdinando Gomes de Freitas ,  Oriol Vinyals ,  Sergio Gomez ,  Ashley Deloris Edwards ,  Jacob Bruce ,  Gabriel Barth-Maron

IPC: G06N3/08 ,  G06N3/04

CPC classification number: G06N3/08 ,  G06N3/04

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting actions to be performed by an agent to interact with an environment using an action selection neural network. In one aspect, a method comprises, at each time step in a sequence of time steps: generating a current representation of a state of a task being performed by the agent in the environment as of the current time step as a sequence of data elements; autoregressively generating a sequence of data elements representing a current action to be performed by the agent at the current time step; and after autoregressively generating the sequence of data elements representing the current action, causing the agent to perform the current action at the current time step.

公开(公告)号：US20230376780A1

公开(公告)日：2023-11-23

申请号：US18029979

申请日：2021-10-01

Applicant: DeepMind Technologies Limited

Inventor： Caglar Gulcehre ,  Razvan Pascanu ,  Sergio Gomez

IPC: G06N3/092 ,  G06N3/0442

CPC classification number: G06N3/092 ,  G06N3/0442

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network used to select actions performed by an agent interacting with an environment by performing actions that cause the environment to transition states. One of the methods includes maintaining a replay memory storing a plurality of transitions; selecting a plurality of transitions from the replay memory; and training the neural network on the plurality of transitions, comprising, for each transition: generating an initial Q value for the transition; determining a scaled Q value for the transition; determining a scaled temporal difference learning target for the transition; determining an error between the scaled temporal difference learning target and the scaled Q value; determining an update to the current values of the Q network parameters; and determining an update to the current value of the scaling term.

公开(公告)号：US20230061411A1

公开(公告)日：2023-03-02

申请号：US17410689

申请日：2021-08-24

Applicant: DeepMind Technologies Limited

Inventor： Tom Erez ,  Alexander Novikov ,  Emilio Parisotto ,  Jack William Rae ,  Konrad Zolna ,  Misha Man Ray Denil ,  Joao Ferdinando Gomes de Freitas ,  Oriol Vinyals ,  Scott Ellison Reed ,  Sergio Gomez ,  Ashley Deloris Edwards ,  Jacob Bruce ,  Gabriel Barth-Maron

IPC: G06N3/08 ,  G06N3/04

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting actions to be performed by an agent to interact with an environment using an action selection neural network. In one aspect, a method comprises, at each time step in a sequence of time steps: generating a current representation of a state of a task being performed by the agent in the environment as of the current time step as a sequence of data elements; autoregressively generating a sequence of data elements representing a current action to be performed by the agent at the current time step; and after autoregressively generating the sequence of data elements representing the current action, causing the agent to perform the current action at the current time step.