Share trained weights of MARL algorithms.

[AmineAndam04]

Hi!
First, congratulations on the outstanding paper.
I was wondering if it would be possible for you to share the final policy weights of the trained MARL models. Doing so would be incredibly helpful for the adversarial MARL community, as it would save us a lot of time that would otherwise be spent training from scratch. This, in turn, would enable us to quickly test and prototype adversarial attacks.

Thanks.