IDSC was honored to present Claus Ekstrøm, professor and vice-chair at the Section of Biostatistics, University of Copenhagen, for a lecture at the Richter Library on Wednesday, December, 1, 2021. Dr. Ekstrøm’s primary research interests are centered on developing methods for the analysis of high-dimensional data problems and causal discovery. Read more “Catch the Replay: Special Guest Lecture on Visual Inference with Claus Ekstrøm”
Join us for a special guest lecture (either in person or virtually) featuring Claus Ekstrøm, professor and vice-chair at the Section of Biostatistics, University of Copenhagen. His primary research interests are centered on developing methods for the analysis of high-dimensional data problems and causal discovery. He’s authored two books on statistics and is frequently used as an expert on statistics in Danish news media. Claus has been a grumpy old man from a young age.
Register Now | Wednesday 12/1/2021, 4:00-5:00 PM
This is a hybrid event: In person at Otto G. Richter Library, CR 343 | or online via Zoom.
Title of the lecture: “Validation of visual inference methods in statistics by use of deep learning”
When does inspecting a certain graphical plot allow for an investigator to reach the right statistical conclusion? Visual inference is commonly used for various tasks in statistics—including model diagnostics and exploratory data analysis – and though attractive due to its intuitive nature, the lack of available methods for validating plots is a major drawback.
We propose a new validation method for visual inference. Our method trains deep neural networks to distinguish between plots simulated under two different data-generating mechanisms (null or alternative), and we use the classification accuracy as a technical validation score (TVS). The TVS measures the information content in the plots, and TVS values can be used to compare different plots or different choices of data-generating mechanisms, thereby providing a meaningful scale that new visual inference procedures can be validated against.
We apply the method to three popular diagnostic plots for linear regression, namely the scatter plot, the quantile-quantile plot, and the residual plot. We consider various types and degrees of misspecification, as well as different within-plot sample sizes. Our method produces TVSs that increase with increasing sample size and decrease with increasing difficulty, and hence the TVS is a meaningful measure of validity.
Join us for a special guest lecture (either in person or virtually) featuring Claus Ekstrøm, professor and vice-chair at the Section of Biostatistics, University of Copenhagen. His primary research interests are centered on developing methods for the analysis of high-dimensional data problems and causal discovery. Read more “Special Guest Lecture on Visual Inference with Claus Ekstrøm 12/1”
You are invited to attend the Sylvester Comprehensive Cancer Center (SCCC) Bioinformatics workshop “Advanced Computing for Biomedical Science—Resources Available to UM Researchers”. Sponsored by the Biostatistics and Bioinformatics Core Facility, the workshop will be held on Friday, September 23, 2016, from 2:30 to 3:30 PM at the Sylvester Support Services Building (SSSB) Auditorium, located at 1430 NW 11th Avenue, Miami 33136 (click for map). Read more “Biomedical Science Resources Workshop 9/23/2016”
Network science is a mature discipline, and in both systems biology and systems medicine applications, offers a variety of methodological tools useful for integrative inference purposes, elucidating in particular molecular interactions of various type and complexity. An advantage that networks offer refers to representation, interpretation, and visualization properties in systems’ embedded modular configurations. The clusters of genes or proteins that one observes grouped into communities, are reducing de facto the degrees of freedom and reorganizing cellular networks according to a mix of localized characteristics reflecting gene—gene and protein—protein spatio-temporal associative dynamics. Read more “Network Science Seminar, Thursday 1/28/2016”
CCS invites you to participate in a Bioinformatics Seminar on Tuesday, June 14th, at 12:00 PM, in the Clinical Research Building, Room 692. Dr. Biju Issac presents, “Microarray Experiment Design: Sample Size Calculations.” Dr. Issac is an Associate Scientist with the Biostatistics and Bioinformatics Core in the Bioinformatics Division at the Sylvester Comprehensive Cancer Center. Read more “Bioinformatics Seminar, Tuesday 6/14/2011”