search

TheVeggydude / MARL-lockdown

A Multi-Agent Reinforcement Learning project that learns to solve the Coordinated Lockdown problem.
MIT License
3 stars 0 forks source link

What is a goal state? #17

Closed TheVeggydude closed 3 years ago

TheVeggydude commented 3 years ago

Our program does not have a clear goal state, because, as opposed to e.g. a maze, we don't have a clear end point.

We have do have some measure of success though, namely the number of infected people.

TheVeggydude commented 3 years ago

We define the goal of the agents to lower the infection amount to at least below some value infection_limit. This could then set a limit for e.g. 10% of the population to be infected.

The reward function should reward achieving this state.