Google DeepMind: The Podcast cover image

Let's get physical

Google DeepMind: The Podcast

00:00

Robotics - Robotic Training

Reinforcement learning works by rewarding a robot when it does what you want. Physical tasks like inserting a key into a lock are subject to the problem known as sparse reward. This might make it seem like reinforcement learning is a magic bullet, but in practice, things are a bit complicated. Some of the agorithms inside a i robot are much like dogs, except they are rewarded with numbers, not tasty biscuits.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app