SPLASH 2020
Sun 15 - Sat 21 November 2020 Online Conference
Thu 19 Nov 2020 15:20 - 15:40 at SPLASH-I - R-5 Chair(s): Anitha Gollamudi, Alex Potanin
Fri 20 Nov 2020 03:20 - 03:40 at SPLASH-I - R-5 Chair(s): Jan Vitek

Recently, there is growing concern that machine-learned software, which currently assists or even automates decision making, reproduces, and in the worst case reinforces, bias present in the training data. The development of tools and techniques for certifying fairness of this software or describing its biases is, therefore, critical. In this paper, we propose a perfectly parallel static analysis for certifying fairness of feed-forward neural networks used for classification of tabular data. When certification succeeds, our approach provides definite guarantees, otherwise, it describes and quantifies the biased input space regions. We design the analysis to be sound, in practice also exact, and configurable in terms of scalability and precision, thereby enabling pay-as-you-go certification. We implement our approach in an open-source tool called Libra and demonstrate its effectiveness on neural networks trained on popular datasets.

Thu 19 Nov

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

15:00 - 16:20
R-5OOPSLA at SPLASH-I +12h
Chair(s): Anitha Gollamudi Harvard University, Alex Potanin Victoria University of Wellington
15:00
20m
Talk
Precise Static Modeling of Ethereum “Memory”
OOPSLA
Sifis Lagouvardos University of Athens, Neville Grech University of Malta, Ilias Tsatiris University of Athens, Yannis Smaragdakis University of Athens
Link to publication DOI Media Attached
15:20
20m
Talk
Perfectly Parallel Fairness Certification of Neural Networks
OOPSLA
Caterina Urban INRIA & École Normale Supérieure | Université PSL, Maria Christakis MPI-SWS, Valentin Wüstholz ConsenSys, Fuyuan Zhang MPI-SWS
Link to publication DOI Media Attached
15:40
20m
Talk
Taming Callbacks for Smart Contract Modularity
OOPSLA
Elvira Albert Complutense University of Madrid, Shelly Grossman Tel Aviv University, Noam Rinetzky Tel Aviv University, Clara Rodríguez-Núñez Complutense University of Madrid, Albert Rubio Complutense University of Madrid, Mooly Sagiv Tel Aviv University
Link to publication DOI Media Attached
16:00
20m
Talk
Exposing Cache Timing Side-Channel Leaks through Out-of-Order Symbolic Execution
OOPSLA
Shengjian Guo Baidu Security, Yueqi Chen Pennsylvania State University, Jiyong Yu University of Illinois at Urbana-Champaign, Meng Wu Ant Group, Zhiqiang Zuo Nanjing University, Peng Li Baidu Security, Yueqiang Cheng Baidu Security, Huibo Wang Baidu Security
Link to publication DOI Media Attached

Fri 20 Nov

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

03:00 - 04:20
R-5OOPSLA at SPLASH-I
Chair(s): Jan Vitek Northeastern University / Czech Technical University
03:00
20m
Talk
Precise Static Modeling of Ethereum “Memory”
OOPSLA
Sifis Lagouvardos University of Athens, Neville Grech University of Malta, Ilias Tsatiris University of Athens, Yannis Smaragdakis University of Athens
Link to publication DOI Media Attached
03:20
20m
Talk
Perfectly Parallel Fairness Certification of Neural Networks
OOPSLA
Caterina Urban INRIA & École Normale Supérieure | Université PSL, Maria Christakis MPI-SWS, Valentin Wüstholz ConsenSys, Fuyuan Zhang MPI-SWS
Link to publication DOI Media Attached
03:40
20m
Talk
Taming Callbacks for Smart Contract Modularity
OOPSLA
Elvira Albert Complutense University of Madrid, Shelly Grossman Tel Aviv University, Noam Rinetzky Tel Aviv University, Clara Rodríguez-Núñez Complutense University of Madrid, Albert Rubio Complutense University of Madrid, Mooly Sagiv Tel Aviv University
Link to publication DOI Media Attached
04:00
20m
Talk
Exposing Cache Timing Side-Channel Leaks through Out-of-Order Symbolic Execution
OOPSLA
Shengjian Guo Baidu Security, Yueqi Chen Pennsylvania State University, Jiyong Yu University of Illinois at Urbana-Champaign, Meng Wu Ant Group, Zhiqiang Zuo Nanjing University, Peng Li Baidu Security, Yueqiang Cheng Baidu Security, Huibo Wang Baidu Security
Link to publication DOI Media Attached