Job Talk #2: Regression Fault Detection and Mitigation in the Evolution of Deep Learning Systems
Deep learning (DL) has been widely used in various industrial practices. Similar to traditional software, DL systems also need to constantly evolve to meet the ever-changing new requirements or fix exposed bugs. However, during the evolution process, new bugs may be induced in functions that were carefully constructed in the previous version. This can result in the originally correct predictions becoming incorrect, and these faults are called regression faults. Regression faults may lead to serious consequences. Therefore, ensuring the quality of DL systems during evolution is important.
Wed 25 JunDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
11:00 - 12:30 | |||
11:00 15mDay opening | Opening Remarks Doctoral Symposium C: Andreas Zeller CISPA Helmholtz Center for Information Security, C: Darko Marinov University of Illinois at Urbana-Champaign, C: Jie M. Zhang King's College London, C: Federica Sarro University College London | ||
11:15 45mKeynote | Research and Impact in Academia and Industry Doctoral Symposium Mark Harman Meta Platforms, Inc. and UCL | ||
12:00 15mTalk | Job Talk #1: Search-based Software Testing Driven by Domain Knowledge Doctoral Symposium | ||
12:15 15mTalk | Job Talk #2: Regression Fault Detection and Mitigation in the Evolution of Deep Learning Systems Doctoral Symposium | ||
Andromeda is located close to the restaurant and the bar, at the end of the corridor on the side of the bar.
From the registration desk, go towards the restaurant, turn left towards the bar, walk until the end of the corridor.
