Code used to run the experiments provided in the ICML submission :
Learning to bid in revenue maximizng auctions.
To rerun the code, we advise to use notebookStrategiesPytorch.ipynb to see how to compute the different strategies in the various settings presented in the paper.
The code uses Python 3.6 and Pytorch 0.4.1.
The networks are defined in bidder_strategy.py
The losses are defined in losses.py
The distributions in distributions.py.
One training can be ran wit the main function in bidder_strategy.py
In utils.py, you can find how the evaluation process is defined.
All the different settings are defined in notebookStrategiesPytorch.ipynb.