arXiv paper updated

[Open Bandit Project]

Dear all,

We have updated the arXiv paper describing the features of our dataset and software: https://arxiv.org/abs/2008.07146

In this update, we clarified the notation and added some advanced OPE estimators (such as Switch, MRDR, and DR shrinkage) to our benchmark experiments.

In addition to the updates, we want to share that our paper was accepted by the following NeurIPS workshops:

• Workshop on Offline Reinforcement Learning: https://offline-rl-neurips.github.io/
• Workshop on Consequential Decision Making in Dynamic Environments: https://dynamicdecisions.github.io/

We are looking forward to meeting you at the virtual conference next month!

Best Regards,
Open Bandit Project Team