Support for Hierarchical RL

[muhd-umer]

Hi! 👋🏻

I'm really impressed by how quickly the rl/ repo has grown. I was wondering if hierarchical reinforcement learning (HRL) is currently supported or can be implemented using existing modules in version 0.5.0.

I did note that HRL is on the roadmap:

...
Our roadmap to get there comprises:
#
# - Distributed solutions
# - Offline RL
# - Greater support for meta-RL
# - Multi-task and hierarchical RL
...

If HRL is under development, I'd be happy to contribute if I have the time.

Thanks,
Umer 💓