03. The Environment - Real World

The Environment

As you've seen, the environment for this project involves controlling a double-jointed arm, to reach target locations.

## Real-World Robotics

Watch this YouTube video to see how some researchers were able to train a similar task on a real robot! The accompanying research paper can be found here.

Training robotic arm to reach target locations in the real world. ([Source](https://www.youtube.com/watch?v=ZVIxt2rt1_4))

Training robotic arm to reach target locations in the real world. (Source)

## Sharing Experience

In the second version of the project environment, there are 20 identical copies of the agent. It has been shown that having multiple copies of the same agent sharing experience can accelerate learning, and you'll discover this for yourself when solving the project!

Sharing experience can accelerate learning. ([Source](https://ai.googleblog.com/2018/06/scalable-deep-reinforcement-learning.html))

Sharing experience can accelerate learning. (Source)