Skip to content
No description, website, or topics provided.
Python Jupyter Notebook
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
exp5_1
exp5_2
README.md

README.md

DoubleReinforcementLearningMDP

This repository contains the code for replicating the experiments from the paper

"Double Reinforcement Learning for Efficient Off-Policy Evaluation in Markov Decision Processes"

Experiments in Section 5.1

The relevant code is in the subdirectory exp5_1.

  • toytoy.py runs the experiment with the in-sample variant of the estimators.
  • toytoy2.py runs the experiment with the samples-splitting variant of the estimators.

For example, to run 10 parallel replications, one can run the command seq 10 | xargs -L 1 -P 10 ./toytoy.sh

Experiments in Section 5.2

The relevant code is in the subdirectory exp5_2.

You can’t perform that action at this time.