BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:iCalendar-Ruby
BEGIN:VEVENT
CATEGORIES:Conferences/Seminars/Lectures
DESCRIPTION:Speaker\;  Miles Cranmer (Princeton)\n\nTitle:  Interpretable M
 achine Learning for physics\n\nAbstract:\n\nWould Kepler have discovered hi
 s laws if machine learning had been around in 1609? Or would he have been s
 atisfied with the accuracy of some black box regression model\, leaving New
 ton without the inspiration to find the law of gravitation? In this talk I 
 will present a review of some industry-oriented machine learning algorithms
 \, and discuss a major issue facing their use in the natural sciences: a la
 ck of interpretability. I will then outline several approaches I have creat
 ed with collaborators to help address these problems\, based largely on a m
 ix of structured deep learning and symbolic methods. This will include an i
 ntroduction to the PySR/SymbolicRegression.jl software (https://astroautoma
 ta.com/PySR)\, a Python/Julia package for high-performance symbolic regress
 ion. I will conclude by demonstrating applications of such techniques and h
 ow we may gain new insights from such results.\n\n         ZOOM Link:  \n\n
                     https://mit.zoom.us/j/99160816852
DTEND:20220921T213000Z
DTSTAMP:20260310T212450Z
DTSTART:20220921T203000Z
LOCATION:MIT-Math Dept.\, /Room 2-449
SEQUENCE:0
SUMMARY:Seminar:  Numerial Methods for Partial Differential Equations
UID:tag:localist.com\,2008:EventInstance_41085580131574
URL:https://calendar.mit.edu/event/numerial_methods_for_partial_differentia
 l_equations
END:VEVENT
END:VCALENDAR