Accepted Papers

Evaluating Off-Policy Evaluation: Sensitivity and Robustness Yuta Saito (Hanjuku-kaso, Co., Ltd.)*; Takuma Udagawa (Sony Corporation); Haruka Kiyohara (Tokyo Institute of Technology); Kazuki Mogi (Stanford University); Yusuke Narita (Yale University); Kei Tateno (Sony Corporation) Poster

Decomposition-Coordination Methods for Finite Horizon Bandit Problems Michel DE LARA (Ecole des Ponts ParisTech); Benjamin Heymann (Criteo)*; Jean-Philippe CHANCELIER (Ecole des Ponts ParisTech)

Smooth Sequential Optimisation with Delayed Feedback Srivas Chennu (Apple)*; Jamie Martin (Apple); Puli Liyanagama (Apple); Phil Mohr (Apple) ArXiv Version (longer)

Variational Causal Networks: Approximate Bayesian Inference over Causal Structures Yashas Annadani (ETH Zurich)*; Jonas Rothfuss (ETH); Alexandre Lacoste (Element AI); Nino Scherrer (ETH Zürich); Anirudh Goyal (University of Montreal); Yoshua Bengio (Mila); Stefan Bauer (MPI IS)

Combining Reward and Rank Signals for Slate Recommendation Imad Aouali (Criteo AI Lab)*; Sergey Ivanov (Criteo); Mike Gartrell (Criteo); David Rohde (Criteo); Flavian Vasile (Criteo); Victor Zaytsev (Criteo); Diego Legrand (Criteo) Poster, Slides

Off-Policy Evaluation with General Logging Policies Yusuke Narita (Yale University); Kyohei Okumura (Northwestern University); Akihiro Shimizu (Mercari); Kohei Yata (Yale University)

Recommendation Using Reward Modelling and Sophisticated Practical Compromises Amine Benhalloum (Criteo); Guillaume Genthial (Criteo); David Rohde (Criteo); Flavian Vasile (Criteo)