[mlpack] Reinforcement Learning Gym

[Gopi Manohar Tatiraju]

Hey,

I saw that for reinforcement learning we are using
*https://github.com/zoq/gym_tcp_api
<https://github.com/zoq/gym_tcp_api>* which provides an infrastructure (TCP
API) to the OpenAI Gym toolkit.

I was wondering if mlpack is planning to implement its own environments?
Like maybe we can start by implementing some Atari Games using C++ and
OpenGL
If we want such a feature then I want to start with making Atari Breakout
and implement it as an example.

Any thoughts?

Regards
Gopi M. Tatiraju
https://github.com/heisenbuug
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://knife.lugatgt.org/pipermail/mlpack/attachments/20200923/5a50b11c/attachment.htm>