Google Reinforcement Learning uses successful examples instead of tricky reward functions
Google Reinforcement Learning uses successful examples instead of tricky reward functions
Google AI blog https://ai.googleblog.com/2021/03/recursive-classification-replacing.html
To teach a robot to hammer a nail into a wall, most reinforcement learning algorithms require that the user define a reward function.
The example-based control method uses examples of what the world looks like when a task is completed to teach the robot to solve the task, e.g., examples where the nail is already hammered into the wall.
Project Web site https://ben-eysenbach.github.io/rce
GitHub https://github.com/google-research/google-research/tree/master/rce
arXiv paper abstract https://arxiv.org/abs/2103.12656
arXiv PDF paper https://arxiv.org/pdf/2103.12656.pdf
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
