| Definition | Main purpose of this entire system is to investigate how human interaction can affect the traditional reinforcement learning framework |
| No of environments | 6 |
The environment leverages the framework as defined by OpenAI Gym to create a custom environment. The environment contains a grid of terrain gradient values. The reward of the environment is predicted coverage, which is calculated as a linear function of the actions taken by the agent.
Main purpose of this entire system is to investigate how human interaction can affect the traditional reinforcement learning framework. Custom scripts were written to facilitate this, and several TensorForce scripts were modified as well. These can be found in the custom scripts folder, which need to be manually extracted and placed in the TensorForce package directory.