Logarithmic Smoothing for Pessimistic Off-Policy Evaluation, Selection and Learning

Open in new window