BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Department of Statistics - ECPv4.9.9//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Department of Statistics
X-ORIGINAL-URL:https://stat.wisc.edu
X-WR-CALDESC:Events for Department of Statistics
BEGIN:VTIMEZONE
TZID:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:20190310T080000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:20191103T070000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20191127T160000
DTEND;TZID=America/Chicago:20191127T170000
DTSTAMP:20191213T085317
CREATED:20191125T191305Z
LAST-MODIFIED:20191125T191315Z
UID:2421-1574870400-1574874000@stat.wisc.edu
SUMMARY:Statistics Seminar
DESCRIPTION:Title: Efficient and Optimal Tensor Regression via Importance Sketching \nPresenter: Anru Zhang \nAbstract: The past decade has seen a large body of work on high-dimensional tensors or multiway arrays that arise in numerous applications. In many of these settings\, the tensor of interest is high-dimensional in that the ambient dimension is substantially larger than the sample size. Oftentimes\, however\, the tensor comes with natural low-rank or sparsity structures. How to exploit such structures of tensors poses new statistical and computational challenges. \nIn this talk\, we develop a novel procedure for low-rank tensor regression\, namely Importance Sketching Low-rank Estimation for Tensors (ISLET)\, to address these challenges. The central idea behind ISLET is what we call importance sketching\, carefully designed sketches based on both the responses and the structures of the parameter of interest. We show that our estimating method is sharply minimax optimal in terms of the mean-squared error under low-rank Tucker assumptions. In addition\, if a tensor is low-rank with group sparsity\, our procedure also achieves minimax optimality. Further\, we show through numerical studies that ISLET achieves comparable mean-squared error performance to existing state-of-the-art methods whilst having substantial storage and run-time advantages. In particular\, our procedure performs reliable tensor estimation with tensors of dimension p = O(10^8) and is 1 or 2 orders of magnitude faster than baseline methods. An application to MRI imaging analysis is finally discussed. \nÂ \n
URL:https://stat.wisc.edu/event/statistics-seminar-3/
LOCATION:133 SMI
CATEGORIES:Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20191204T160000
DTEND;TZID=America/Chicago:20191204T170000
DTSTAMP:20191213T085317
CREATED:20190724T193358Z
LAST-MODIFIED:20191202T144034Z
UID:1766-1575475200-1575478800@stat.wisc.edu
SUMMARY:Statistics Seminar
DESCRIPTION:Title: Statistical analysis for Markov-dependent data \nPresenter: Stephen Berg \nAbstract: In the first part of the talk\, I will discuss modeling the species distribution of trees in Wisconsinâ€™s historical forests\, based on data collected from the Wisconsin Public Land Survey (PLS). Our approach uses a finite mixture model with a Markov random field distribution for the latent labels in order to identify and map forest community subtypes\, while accounting for spatial correlation. I will introduce a stochastic approximation maximum likelihood method using Markov chain Monte Carlo (MCMC) for estimating model parameters and discuss the resulting community maps for the PLS data. In the second part of the talk\, I will introduce a novel control variate method for increasing the efficiency of MCMC simulations involving deterministic sweep Markov chains. I will demonstrate that the proposed control variate estimator is easy to implement and outperforms the existing methodology both theoretically and empirically for arbitrary two-component Gibbs samplers. \n
URL:https://stat.wisc.edu/event/statistics-seminar-2019-12-04/
LOCATION:140 Bardeen
CATEGORIES:Seminar
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20191206T160000
DTEND;TZID=America/Chicago:20191206T160000
DTSTAMP:20191213T085317
CREATED:20190904T192307Z
LAST-MODIFIED:20191025T160536Z
UID:1904-1575648000-1575648000@stat.wisc.edu
SUMMARY:Department Happy Hour
DESCRIPTION:
URL:https://stat.wisc.edu/event/department-happy-hour-2-2-2/
LOCATION:The Sett Union South\, 1308 W Dayton St room 105\, Madison\, WI\, 53715\, United States
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=America/Chicago:20191211T160000
DTEND;TZID=America/Chicago:20191211T170000
DTSTAMP:20191213T085317
CREATED:20190807T144314Z
LAST-MODIFIED:20191203T142121Z
UID:1813-1576080000-1576083600@stat.wisc.edu
SUMMARY:Statistics Seminar
DESCRIPTION:Title: Statistical Inference for Large-Scale Data with Incomplete Labels \nPresenter: Hyebin Song \nAbstract: In various real-world problems\, we are presented with data with partially observed or contaminated labels. One example is datasets from deep mutational scanning (DMS) experiments in proteomics\, which typically do not contain non-functional sequences. This talk addresses statistical inference procedures for analyzing noisy\, high-dimensional binary data. In the first part of the talk\, I will discuss variable selection in the context of positive-unlabeled data when the number of features p is large. I present the PUlasso algorithm for variable selection and classification with positive and unlabeled responses\, which is scalable to large-scale data and equipped with the minimax optimal mean-squared error guarantee. In the second part of the talk\, I will discuss statistical inference procedures with noisy labels data. With the key observation that the noisy labels problem belongs to a special sub-class of generalized linear models\, I will present convex and non-convex approaches for inference with statistical guarantees. Finally\, I will present an application of our methodology to inferring sequence-function relationships and designing highly stabilized enzymes from large-scale DMS data. \n
URL:https://stat.wisc.edu/event/statistics-seminar-2/
LOCATION:140 Bardeen
CATEGORIES:Seminar
END:VEVENT
END:VCALENDAR