Training action selection neural networks using leave-one-out-updates
US11604997B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Jun 11, 2018 |
| Grant date | Mar 14, 2023 |
| Priority date | — |
| Expiry date | Feb 23, 2040 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N3/092
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a policy neural network. The policy neural network is used to select actions to be performed by an agent that interacts with an environment by receiving an observation characterizing a state of the environment and performing an action from a set of actions in response to the received observation. A trajectory is obtained from a replay memory, and a final update to current values of the policy network parameters is determined for each training observation in the trajectory. The final updates to the current values of the policy network parameters are determined from selected action updates and leave-one-out updates.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.