Official implementation for the paper More Benefits of Being Distributional: Second-Order Bounds for Reinforcement Learning.
We provide code for regcb and the distUCB algorithm and toggle between the two using the alg hyperparameter. The hyperparameters are found in conf with the main being config.yaml.
One can install the required packages by running pip install -r requirements.txt.
Both the prudential and housing datasets require preprocessing. To do this, one can run prudential_preprocessing.py and king_housing_preprocessing.py respectively. To download the prudential dataset, one should look at the Prudential Kaggle competition and use that CSV (https://www.kaggle.com/c/prudential-life-insurance-assessment). Preprocessing this dataset takes a bit of time. The king housing arff dataset is provided under datasets/dataset.arff
To run the code, one can use the following command:
accelerate launch main.py