BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Debugging Machine Learning Code - Svet Penkov\, Efemarai
DTSTART:20210119T131500Z
DTEND:20210119T141500Z
UID:TALK155641@talks.cam.ac.uk
CONTACT:Mateja Jamnik
DESCRIPTION:"Join us on Zoom":https://zoom.us/j/99166955895?pwd=SzI0M3pMVE
 kvNmw3Q0dqNDVRalZvdz09\n\nDeveloping robust machine learning code is hard\
 , but debugging it is even harder! When developing machine learning (ML) s
 olutions for complex problems such as autonomous driving we tend to focus 
 on the hard research problems and develop theoretical models whose impleme
 ntation we take for granted. In practice\, however\, bugs always creep in.
  How do you detect bugs when working with multidimensional arrays containi
 ng millions of parameters? How do you identify sources of error when build
 ing dynamic computational graphs? In this talk I will provide an overview 
 of the existing methods for debugging ML code and showcase the first of it
 s kind visual 3D debugger which makes debugging deep learning models a bre
 eze.\n\nBio: \nAfter finishing his PhD in AI & Robotics at the University 
 of Edinburgh\, Dr. Penkov led the motion prediction team at Five AI for mo
 re than 2 years. Currently\, he is the CEO & co-founder of Efemarai whose 
 mission is to bring software QA to machine learning.\n
LOCATION:Zoom
END:VEVENT
END:VCALENDAR