lola

lola copied to clipboard

Pervasive reshape bugs in train_cg?

1

Unless I misunderstand what the code is trying to do, the following pattern in `train_cg.py` is a bug: ```python trainBatch1 = [[], [], [], [], [], []] # line 137...

cooijmanstim

Are the current default hyper-parameters the one used to produce the results of the [DICE paper](https://arxiv.org/pdf/1802.05098.pdf)? Current default HP are (from [scripts/run_lola_dice.py](https://github.com/alshedivat/lola/blob/master/scripts/run_lola_dice.py)): ``` batch-size=64 runs=5 epochs=200 use_dice=True gamma=.96, lr_inner=.1, lr_outer=.2,...

Manuscrit

In the notebook [notebooks/dice/analysis.ipynb](https://github.com/alshedivat/lola/blob/master/notebooks/dice/analysis.ipynb) which is used to analyse the results and reproduce the fig.5 from the paper [DiCE: The Infinitely Differentiable Monte Carlo Estimator](https://arxiv.org/pdf/1802.05098.pdf), the confidence interval used is...

Manuscrit

Player blue and red are not currently symmetrical

1

In https://github.com/alshedivat/lola/blob/master/lola/envs/coin_game.py The symmetry is broken in favor of player red. When the two players move at the same time on the cell with the coin, player red has the...

Manuscrit

Hello, thank you for open-sourcing the code! :-) The code is really helpful in understanding the papers deeper. I am interested in LOLA, especially its policy gradient method ([lola/train_pg.py](https://github.com/alshedivat/lola/blob/master/lola/train_pg.py)). As...

dkkim93

Coin Game

7

Can you suggest a sample command line to run Coin Game? I tried running just: python scripts/run_lola.py --exp_name=CoinGame --no-exact and it seems to be updating parameters and using up all...

LUKELIEM

LOLA breaks when changing number of actions and/or states

5

I try to edit IPD to a setup with four actions. This yields a 4x4 payoff matrix and a 17-dimensional input, which breaks both LOLA and LOLA-DiCE implementations. - train_exact.py...

jazzbob

In coin game the input shape is [4,grid_size,grid_size]. Shouldn't the input be transposed before passing to the model in networks.py so that the channel(4) becomes the last dimension?

a7b23