Marginal Density Ratio for Off-Policy Evaluation in Contextual Bandits.
Muhammad Faaiz TaufiqArnaud DoucetRob CornishJean-Francois TonPublished in: CoRR (2023)
Keyphrases
- density ratio
- policy evaluation
- least squares
- semi parametric
- density ratio estimation
- density estimation
- linear model
- regression model
- linear regression
- statistical inference
- probability distribution
- policy iteration
- optical flow
- constrained optimization
- parametric models
- objective function
- gaussian mixture model
- data mining
- model free
- computer vision