BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.4.5//EN TZID:America/Los_Angeles X-WR-TIMEZONE:America/Los_Angeles BEGIN:VEVENT UID:85@escience.washington.edu DTSTART;TZID=America/Los_Angeles:20240417T163000 DTEND;TZID=America/Los_Angeles:20240417T172000 DTSTAMP:20240415T162612Z URL:https://escience.washington.edu/events/uw-data-science-seminar-armeen- taeb/ SUMMARY:UW Data Science Seminar: Armeen Taeb DESCRIPTION:\n\nPlease join us for a UW Data Science Seminar on Wednesday\, April 17th from 4:30 to 5:20 p.m. PST. The seminar will feature Armeen Ta eb\, an Assistant Professor in the UW Department of Statistics.\n\nThe sem inar will be held in the Physics/Astronomy Auditorium (PAA)\, Room A118 -  campus map.\n\n\n \;\n"Model Selection and False Positive Error Cont rol in Complex Modeling Paradigms"\nAbstract: Controlling the false positi ve error in model selection is a prominent paradigm for gathering evidence in data-driven science.  In model selection problems such as variable se lection and graph estimation\, models are characterized by an underlying B oolean structure such as presence or absence of a variable or an edge.  T herefore\, false positive error or false negative error can be convenientl y specified as the number of variables/edges that are incorrectly included or excluded in an estimated model.  However\, the increasing complexity of modern datasets has been accompanied by the use of sophisticated modeli ng paradigms in which defining false positive error is a significant chall enge.  For example\, models specified by structures such as partitions (f or clustering)\, permutations (for ranking)\, directed acyclic graphs (for causal inference)\, or subspaces (for principal components analysis) are not characterized by a simple Boolean logical structure\, which leads to d ifficulties with formalizing and controlling false positive error.  We pr esent a generic approach to endow a collection of models with partial orde r structure\, which leads to systematic approaches for defining natural ge neralizations of false positive error and methodology for controlling this error.  (Joint work with Peter Bühlmann\, Venkat Chandrasekaran\, and P arikshit Shah)\nBio: Armeen Taeb is an assistant professor in the Departme nt of Statistics at the University of Washington. His research interests l ie at the interface of optimization and statistics. His work currently foc uses on developing efficient methods for graphical and latent-variable mod eling\, learning provably optimal causal models from data\, domain adaptat ion\, and false positive error control in non-traditional settings. He is also interested in exploring the utility of statistical methodologies for real-world applications\, especially in the earth sciences. Prior to comin g to UW\, Armeen was a postdoctoral fellow of ETH Foundations of Data Sci ence (ETH-FDS) at ETH Zürich\, mentored by Peter Bühlmann. Previously\ , under the supervision of Venkat Chandrasekaran\, he obtained my PhD in the Department of Electrical Engineering at Caltech.\n\n\nThe UW Data Scie nce Seminar is an annual lecture series at the University of Washington t hat hosts scholars working across applied areas of data science\, such as the sciences\, engineering\, humanities and arts along with methodological areas in data science\, such as computer science\, applied math and stati stics. Our presenters come from all domain fields and include occasional e xternal speakers from regional partners\, governmental agencies and indust ry.\nThe 2023-2024 seminars will be held in person\, and are free and open to the public. LOCATION:Physics/Astronomy Auditorium\, room A118\, Seattle\, WA\, X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Seattle\, WA\, ;X-APPLE-RAD IUS=100;X-TITLE=Physics/Astronomy Auditorium\, room A118:geo:0,0 END:VEVENT BEGIN:VTIMEZONE TZID:America/Los_Angeles X-LIC-LOCATION:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20240310T030000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 TZNAME:PDT END:DAYLIGHT END:VTIMEZONE END:VCALENDAR