Open RiccZamboni opened 1 year ago
Hey @RiccarDigno , thanks for raising this request. We are indeed working on a solution for this problem right now via our new and upcoming RLModule (and RLTrainer) APIs. These will enable you to fully customize shared multi-agent sub-models (such as a comm. channel) and plug these into any of our existing algorithms.
For a first glimpse, see here: https://github.com/ray-project/ray/tree/master/rllib/core/rl_module
Assigning @kouroshHakha , who is in charge of this effort.
Hey @sven1977, thank you I will definitely have a look!
Hi, I'm a bot from the Ray team :)
To help human contributors to focus on more relevant issues, I will automatically add the stale label to issues that have had no activity for more than 4 months.
If there is no further activity in the 14 days, the issue will be closed!
You can always ask for help on our discussion forum or Ray's public slack channel.
Hi again! The issue will be closed because there has been no more activity in the 14 days since the last message.
Please feel free to reopen or open a new issue if you'd still like it to be addressed.
Again, you can always ask for help on our discussion forum or Ray's public slack channel.
Thanks again for opening the issue!
Hi, I'm a bot from the Ray team :)
To help human contributors to focus on more relevant issues, I will automatically add the stale label to issues that have had no activity for more than 4 months.
If there is no further activity in the 14 days, the issue will be closed!
You can always ask for help on our discussion forum or Ray's public slack channel.
Description
Currently, no information sharing among agents is available, out of sharing the same policy and/or having centralized information among them. As the number of algorithms in MARL exploiting these techniques keeps growing, the inclusion of some communication channels between agents seems to be a nice-to-have feature
Use case
Any MARL algorithm belonging neither to the independent learning setting nor to the centralized training one requires for scalable and flexible communication between agents