Optimal and Adaptive Off-policy Evaluation in Contextual Bandits

2017

Subfields: Reinforcement Learning

Citations in the corpus

• Policy Evaluation and Optimization with Continuous Treatments (2018, background): “Recent work in [23, 24] switches between using an IPS and reward estimator, using the reward estimator when the propensity is smaller than some threshold which optimizes the MSE bias-variance trade-off.”
• What are the Statistical Limits of Offline RL with Linear Function Approximation? (2020, unknown): “…[Dud´ık et al., 2011, Mandel et al., 2014, Thomas et al., 2015, Li et al., 2015, Jiang and Li, 2016, Thomas and Brunskill, 2016, Guo et al., 2017, Wang et al., 2017, Liu et al., 2018, Farajtabar et al., 2018, Xie et al., 2019, Kallus and Uehara, 2019, Liu et al., 2019, Uehara and Jiang, 2019,…”
• Imitation-Regularized Offline Learning (2019, methodology): “On the other hand, we extend our methods to doubly robust approaches [25, 24, 2, 13, 8] and switching approaches [33, 18, 32] for additional, free variance reduction.”
• Instabilities of Offline RL with Pre-Trained Neural Representation (2021, unknown): “…[Dud´ık et al., 2011, Mandel et al., 2014, Thomas et al., 2015, Li et al., 2015, Jiang and Li, 2016, Thomas and Brunskill, 2016, Guo et al., 2017, Wang et al., 2017, Liu et al., 2018, Farajtabar et al., 2018, Xie et al., 2019, Kallus and Uehara, 2019, Liu et al., 2019, Uehara and Jiang, 2019,…”