BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Learning Program Properties From Big Code For Software Fault Local
 ization
DTSTART:20160718T130000
DTEND:20160718T150000
DTSTAMP:20260408T060107Z
UID:284aedb0bc2846853b9b79a31458ee216b1664e8efd818c642a624e6
CATEGORIES:Conferences - Seminars
DESCRIPTION:David Aksun\nEDIC Candidacy Exam\nExam President: Prof. Viktor
  Kuncak\nThesis Director: Prof. James Larus\nCo-examiner: Prof. Katerina A
 rgyraki\nBackground papers:Finding latent code errors via machine learning
  over program executions\, by Y. Brun\, M. D. Ernst.Predicting Program Pro
 perties from "Big Code\, by V. Raychev et al.Automatic Patch Generation by
  Learning Correct Code\, by F. Long\, M. Rinard.Abstract\nAutomated softwa
 re fault localization systems are essential for identifying the locations 
 of software faults. Most of the automated tools require explicit program s
 pecifications\, such as test suites and formal specifications. We can infe
 r specifications for software fault localization using the information fro
 m bug fixes and get these bug fixes from large open source repositories (b
 ig code).\nRecently\, tools based on probabilistic models of code trained 
 from big code are scalable\, effective and provide significant results in 
 areas\, such as code completion\, fault finding\, code beautification\, co
 de language translation and mining code patterns. In our work\, we can use
  these probabilistic models to find specifications from bug fixes to locat
 e program statements that are likely to be repaired.\nIn this research pro
 posal\, we investigate a machine learning formulation of software fault lo
 calization based on bug fixes. We examine an expressive\, scalable probabi
 listic model of code\, which can learn program properties from complex dep
 endencies. Then\, we present a program repair system\, which ranks candida
 te patches based on a model of correct code. Finally\, we propose building
  software fault localization tools based on probabilistic models of code u
 sing bug fixes.
LOCATION:BC 410 https://plan.epfl.ch/?room==BC%20410
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR