[CFP] Deadline 7 October, NeurIPS 2021 Causal Inference Challenges in Sequential Decision Making

[Masatoshi Uehara]

Call for papers

NeurIPS 2021 Causal Inference Challenges in Sequential DecisionMaking: Bridging Theory and Practice

website: https://sites.google.com/view/causal-sequential-decisions/home

Submission deadline: October 7th, 2021 (Anywhere on Earth ) (Deadline is extended)

 

Workshop overview

Sequential decision-making problems appear in settings as varied as healthcare, e-commerce, operations management, and policymaking, and depending on the context these can have very varied features that make each problem unique, whether they involve online learning or offline data, known cost structures or unknown counterfactuals, continuous actions with or without constraints or finite or combinatorial actions, stationary environments or environments with dynamic agents, utilitarian considerations or fairness or equity considerations. More and more, causal inference and discovery and adjacent statistical theories have come to bear on such problems, from the early work on longitudinal causal inference from the last millenium up to recent developments in bandit algorithms and inference, dynamic treatment regimes, both online and offline reinforcement learning, interventions in general causal graphs and discovery thereof, and more. While the interaction between these theories has grown, expertise is spread across many different disciplines, including CS/ML, (bio)statistics, econometrics, ethics/law, electrical engineering, and operations research. 

Topics that the workshop will seek invited and contributed talks on include but not limited to:

- Dynamic treatment regimes

- Causal inference and discovery from longitudinal and panel data

- Unmeasured confounding in sequential decisions and sensitivity analyses

- Inference from adaptive (bandit) experiments

- Causality in dynamical systems

- Online/offline A/B testing

- Econometric/structural estimation in sequential settings

- Offline/Online reinforcement learning and bandits

- Algorithmic fairness in dynamic environments

- Causal graphs with multiple and/or sequential interventions

- Online allocation and online linear programs

- Data-driven inverse optimization in sequential settings

- Applications, including in healthcare, e-commerce, and policymaking

 

Submission  Submissions should use the NeurIPS 2021 template, and be 2-6 pages (plus as many pages as necessary for references and supplementary material). Submissions of extended abstracts of longer work is allowed and, after acceptance, these may directly link to the longer version on a preprint server (e.g., on arXiv). Submissions should otherwise be anonymized during the review process. We remark that looking at supplementary material is at the discretion of the reviewers. All submissions will be peer-reviewed. There are no official archival proceedings. Contributors can still publish accepted work in archival journals or conferences. The submissions will only be posted on the workshop website for presentation purposes. Authors will have the opportunity to link to a version on a preprint server. A subset of submissions will be invited for a spotlight talk, and the remaining will be invited to present at the poster session.

Deadlines and Dates

· Submission deadline: October 7th, 2021 (Anywhere on Earth )

· Notification: 23 October 2021 (Anywhere on Earth)

· Workshop (virtual): 14 December 2021

Organizers

Aurélien Bibaut (Netflix)

Maria Dimakopoulou (Netflix)

Nathan Kallus (Cornell)

Xinkun Nie (Stanford)

Masatoshi Uehara (Cornell)

Kelly Zhang (Harvard)