Marc Lanctot

Hi @StepHaze, It's probably due to the size of Mancala? i.e. I would guess something similar would happen with chess. The state space is too large, so most states are...

Sure. In OpenSpiel all the agents follow the same API (they are subclasses of the agent base class). So, 99% of the code above can be re-used and you just...

Note the example above is a bit old and still uses the TF1-based DQN. It can be easily swapped for the [PyTorch DQN](https://github.com/google-deepmind/open_spiel/blob/master/open_spiel/python/pytorch/dqn.py) or [JAX DQN](https://github.com/google-deepmind/open_spiel/blob/master/open_spiel/python/jax/dqn.py) if you prefer.

Sounds like you are talking about checkpoints. The DQN supports saving and loading the networks themselves, but not the replay buffer. But it's certainly not enabled by default. It's something...

Hi @giogix2, thanks! First thing, can you sign the CLA? (See the "Details" link above and follow the instructions.)

> My Question now is whether i can use newer CUDA drivers for example 12.2 with latest Cudnn. In the global_variables.sh i could only see the options 10.2 and lower...

@TheSQLGuru I agree, that would be great. Will require some community coordination. @CasparQuast, are you willing to post your code up somewhere in a fork or pull request (which creates...

> I've been playing with open_spiel's R-NAD algorithm implementation in python and noticed some strange behavior: each time R-NAD calls `state.observation_tensor()` a new state is created, then there is a...

Ok I think this question is more about how the Python games work than about R-NaD. The code for those functions is here: https://github.com/deepmind/open_spiel/blob/master/open_spiel/python/pybind11/python_games.cc It is possible that there are...

Ok perfect, thanks for the detail. I will check with @elkhrt to see if this was known. Would be nice if we could avoid this.