Fooling robots and building better ones: adversarial ML attacks and defences
Like Add Share
Cybersecurity • May 2021
Fooling robots and building better ones: adversarial ML attacks and defences
Like Add Share
About
Fooling robots and building better ones: adversarial ML attacks and defences
About

In this talk, we'll go over part of the landscape of adversarial machine learning from both perspectives. First we'll explore how testers can attack different types of models, and what an organisation stands to lose. After that we'll dive into both operational and technical mitigations a data scientist or architect can undertake to avoid falling prey to these attacks.

Language
English
Level
Intermediate
Length
36 minutes
Type
online conference
About the speaker
About the speaker
Alwin Peppels
Security SpecialistCyberSeals
I'm a security researcher, working mainly as a consultant for financial and governmental organisations. I have done some of Responsible Disclosures in my free time, and I'm grateful that Microsoft, Philips and Google have given me a spot in their respective Halls of Fame for it.
Details
Language
English
Level
Intermediate
Length
36 minutes
Type
online conference