I am creating some multi-agent RL environments, as mentioned in the title, and got interesting results for the crowd and predator-prey simulations where in the latter, there is a game-theoretic relationship in-between.
I am also working on more interesting self-play army simulations with the same logic. I am writing to invite you to collaborate as I am working alone on this hobby project of mine, and maybe integrating them to evotorch.
(I am also planning to finish a competitive multi-agent closed-market auction environment later.)
Hello again,
I hope that you're doing well.
I am creating some multi-agent RL environments, as mentioned in the title, and got interesting results for the crowd and predator-prey simulations where in the latter, there is a game-theoretic relationship in-between.
Here is the outcome of predator-prey: https://www.linkedin.com/feed/update/urn:li:activity:7023631787763355648/ The source codes for those various simulations are available here: https://github.com/kayuksel/multi-rl-crowd-sim
I am also working on more interesting self-play army simulations with the same logic. I am writing to invite you to collaborate as I am working alone on this hobby project of mine, and maybe integrating them to evotorch.
(I am also planning to finish a competitive multi-agent closed-market auction environment later.)
Sincerely, Kamer