Tue 14 Jun 2022 11:45 - 12:10 at Boardroom - Paper Session 1 Chair(s): Caterina Urban
Tue 14 Jun 2022 23:45 - 00:10 at Boardroom - Paper Session 1

Data leakage is a well-known problem in machine learning. Data leakage occurs when information from outside the training dataset is used to create a model. This phenomenon renders a model excessively optimistic or even useless in the real world, since the model tends to leverage greatly on the unfairly acquired information. To date, detection of data leakages occurs most-mortem using runtime methods. In this paper, we develop a static data leakage analysis to detect several instances of data leakages during development time. Our analysis is constructed to be light weight so that it can be performed in seconds. We have integrated our analysis into the NBLyzer static analyzer. To the best of our knowledge, we propose the first static detection of data leakages.

Tue 14 Jun

Displayed time zone: Pacific Time (US & Canada) change

10:30 - 12:00
Paper Session 1 SOAP at Boardroom +12h
Chair(s): Caterina Urban INRIA & École Normale Supérieure | Université PSL

All papers will be allocated a time slot of 25 min (20min talk + 5 min questions)

10:30
25m
Talk
Abstract interpretation of Michelson smart-contracts
SOAP
P: Guillaume Bau , Antoine Miné Sorbonne Université, Vincent Botbol Nomadic Labs, Mehdi Bouaziz Nomadic Labs Paris
10:55
25m
Talk
BinFPE: Accurate Floating-Point Exception Detection for GPU Applications
SOAP
P: Ignacio Laguna Lawrence Livermore National Laboratory, Xinyi Li University of Utah, Ganesh Gopalakrishnan University of Utah
11:20
25m
Talk
Towards an Implementation of Differential Dynamic Logic in PVSvirtual
SOAP
P: J Tanner Slagel , César Muñoz NASA Langley Research Center, Swee Balachandran National Institute of Aerospace, Mariano Moscato National Institute of Aerospace, Aaron Dutle NASA Langley Research Center, Paolo Masci National Institute of Aerospace, USA, Lauren White NASA Langley Research Center
11:45
25m
Talk
Statically Detecting Data Leakages in Data Science Codevirtual
SOAP
Pavle Subotic Microsoft Azure, Uros Bojanic Microsoft, Milan Stojic Microsoft
22:30 - 00:00
Paper Session 1 SOAP at Boardroom
22:30
25m
Talk
Abstract interpretation of Michelson smart-contracts
SOAP
P: Guillaume Bau , Antoine Miné Sorbonne Université, Vincent Botbol Nomadic Labs, Mehdi Bouaziz Nomadic Labs Paris
22:55
25m
Talk
BinFPE: Accurate Floating-Point Exception Detection for GPU Applications
SOAP
P: Ignacio Laguna Lawrence Livermore National Laboratory, Xinyi Li University of Utah, Ganesh Gopalakrishnan University of Utah
23:20
25m
Talk
Towards an Implementation of Differential Dynamic Logic in PVSvirtual
SOAP
P: J Tanner Slagel , César Muñoz NASA Langley Research Center, Swee Balachandran National Institute of Aerospace, Mariano Moscato National Institute of Aerospace, Aaron Dutle NASA Langley Research Center, Paolo Masci National Institute of Aerospace, USA, Lauren White NASA Langley Research Center
23:45
25m
Talk
Statically Detecting Data Leakages in Data Science Codevirtual
SOAP
Pavle Subotic Microsoft Azure, Uros Bojanic Microsoft, Milan Stojic Microsoft