ML agents on difficulty adjusting

I’m doing my thesis work on reinforcement learning and dynamic difficulty adjusting. I want to use ml agents for it. my main idea is creating an agent which will tune different variables(like the enemies attack, speed, etc) according to the current state of the player. Has anyone done something similar? how would i go through with it?

Are you familiar with PCGRL? Game Level Design with Reinforcement Learning | by Chintan Trivedi | deepgamingai | Medium
It’s probably not exactly what you’re looking for, but maybe it can give you some clues.

I’ll give it a look. Thank you very much!

Hello i know this is an old post but i myself am doing a thesis on the same topic this year and found this page. Were you able to successfully implement DDA in Ml Agents?

1 Like