Towards Defenses for Data Poisoning Attacks
Speaker
Matthew Jagielski
Northeastern University
Host
Srinivas Devadas
Abstract:
As machine learning applications leverage increasingly large datasets, it becomes easier to add malicious data into training datasets. Called data poisoning attacks, these can compromise the learned model in several ways. Defending against these attacks is important for the reliable deployment of machine learning. In this talk, we discuss two papers which design improved poisoning attacks and defenses. In the process, we will also discuss poisoning attacks more broadly and highlight some open questions for defending against them.
Zoom Info:
Join Zoom Meeting
https://mit.zoom.us/j/97527284254
Password: <3security
One tap mobile
+16465588656,,97527284254# US (New York)
+16699006833,,97527284254# US (San Jose)
Meeting ID: 975 2728 4254
US : +1 646 558 8656 or +1 669 900 6833
International Numbers: https://mit.zoom.us/u/ackqPS3AP4
Join by SIP
97527284254@zoomcrc.com
Join by Skype for Business
https://mit.zoom.us/skype/97527284254
As machine learning applications leverage increasingly large datasets, it becomes easier to add malicious data into training datasets. Called data poisoning attacks, these can compromise the learned model in several ways. Defending against these attacks is important for the reliable deployment of machine learning. In this talk, we discuss two papers which design improved poisoning attacks and defenses. In the process, we will also discuss poisoning attacks more broadly and highlight some open questions for defending against them.
Zoom Info:
Join Zoom Meeting
https://mit.zoom.us/j/97527284254
Password: <3security
One tap mobile
+16465588656,,97527284254# US (New York)
+16699006833,,97527284254# US (San Jose)
Meeting ID: 975 2728 4254
US : +1 646 558 8656 or +1 669 900 6833
International Numbers: https://mit.zoom.us/u/ackqPS3AP4
Join by SIP
97527284254@zoomcrc.com
Join by Skype for Business
https://mit.zoom.us/skype/97527284254