Fifth Workshop on Higher-Order Asymptotics and Post-Selection Inference; June 21-23, 2020; St. Louis, Missouri

该奖项支持研究生，博士后和其他早期职业研究人员参加高阶渐近和后选择推理（WHOA-PSI）的第五次研讨会，将于2020年6月21日至6月23日在圣路易斯华盛顿大学举行。讲习班的目标包括：（一）提供一个论坛，传播和讨论模型选择或降维后推理的最新进展;（二）促进理论，方法和应用研究人员之间的合作;和（三）宣传新的选择后推理方法的应用程序是必要的。研讨会包括30多个特邀演讲和2个海报会议，代表了来自世界各地在相关领域工作的大量研究小组。 在过去一年中，新的实质性活动领域包括因果推理，采用机器学习方法后选择性推理的可行性，以及贝叶斯选择后推理。线性模型中的后选择推理仍然是一个非常活跃的研究领域，对于什么应该是推理的目标，控制哪些错误，应该使用什么样的条件，以及如何结合联合收割机统计原理，如决策理论最优性，Fisherian相关性，鲁棒性和Neyman-Pearson方法来测试，尚未达成共识。我们的理解，何时以及如何适当的引导方法可以在选择后推理和高维推理的背景下发展继续发展。这次研讨会为进一步讨论这些重要问题并取得进展提供了机会。它还将提高人们对需要新推理工具的应用问题的认识。该研讨会的网站是https://www.math.wustl.edu/~kuffner/WHOA-PSI-5.html.This奖反映了NSF的法定使命，并已被认为值得通过使用基金会的知识价值和更广泛的影响审查标准进行评估的支持。

This award supports participation of graduate students, postdocs and other early-career researchers in the Fifth Workshop on Higher-Order Asymptotics and Post-Selection Inference (WHOA-PSI), to take place June 21 through June 23, 2020 at Washington University in St. Louis. The goals of the workshop include (i) to provide a forum to disseminate and discuss the most recent advances in inference after model selection or dimension reduction; (ii) to promote collaboration between theoretical, methodological and applied researchers; and (iii) to advertise applications for which new post-selection inference methods are needed. The workshop features more than 30 invited talks and 2 poster sessions, representing a large number of research groups who work in relevant fields from all over the world. New areas of substantial activity within the past year include causal inference, the feasibility of selective inference after machine learning methods have been employed, and Bayesian post-selection inference. Post-selection inference in linear models remains a very active research area, with no consensus yet reached regarding what should be the targets of inference, which errors to control, what sort of conditioning should be used, and how to combine statistical principles such as decision-theoretic optimality, Fisherian relevance, robustness, and Neyman-Pearson approaches to testing. Our understanding of when and how appropriate bootstrap methods can be developed in the context of post-selection inference and high-dimensional inference continues to evolve. This workshop provides an opportunity for further discussion and progress on these important issues. It will also increase awareness of applied problems in need of new inference tools. The website for this workshop is https://www.math.wustl.edu/~kuffner/WHOA-PSI-5.html.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.