Patent · US Active

Reinforcement learning using agent curricula

US11113605B2 · kind B2 · utility

15Cited by

0References

17Claims

0Family size

Assignee

DeepMind Technologies Limited · GB

Inventors

Wojciech Czarnecki · London, GB
Siddhant Jayakumar · London, GB

Key dates

Filing date	May 20, 2019
Grant date	Sep 7, 2021
Priority date	—
Expiry date	May 20, 2039

Classification

Technology area (CPC G)Physics
CPC primaryG06N3/086
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for reinforcement learning using agent curricula. One of the methods includes maintaining data specifying plurality of candidate agent policy neural networks; initializing mixing data that assigns a respective weight to each of the candidate agent policy neural networks; training the candidate agent policy neural networks using a reinforcement learning technique to generate combined action selection policies that result in improved performance on a reinforcement learning task; and during the training, repeatedly adjusting the weights in the mixing data to favor higher-performing candidate agent policy neural networks.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.