No
Yes
View More
View Less
Working...
Close
OK
Cancel
Confirm
System Message
Delete
Schedule
An unknown error has occurred and your request could not be completed. Please contact support.
Scheduled
Wait Listed
Personal Calendar
Speaking
Conference Event
Meeting
Interest
There aren't any available sessions at this time.
Conflict Found
This session is already scheduled at another time. Would you like to...
Loading...
Please enter a maximum of {0} characters.
{0} remaining of {1} character maximum.
Please enter a maximum of {0} words.
{0} remaining of {1} word maximum.
must be 50 characters or less.
must be 40 characters or less.
Session Summary
We were unable to load the map image.
This has not yet been assigned to a map.
Search Catalog
Reply
Replies ()
Search
New Post
Microblog
Microblog Thread
Post Reply
Post
Your session timed out.
This web page is not optimized for viewing on a mobile device. Visit this site in a desktop browser to access the full set of features.
2018 GTC Washington DC

DC8114 - Resisting Adversarial Attacks on Machine Learning Malware Detectors

Session Speakers
Session Description

We will present a simple technique showing how to make these specific machine learning models immune to white-box gradient based attacks at a small cost to accuracy. This makes our solution robust to numerous attacks, and could benefit other domains with binary targeted attacks. Deep Learning is becoming a popular component to building malware detection systems. This is a domain where we need to be particularly concerned with adversarial attacks, which can be done easily today. We have looked at machine learning approaches which require no domain knowledge to train, such as MalConv, and their susceptibility to adversarial attack. Through empirical evaluation, we have found that this machine learning based solution can be more robust to black-box attacks that defeat anti-virus engines today.


Additional Information
Cyber Security
Deep Learning and AI, Cyber Security
General, Defense, Government / National Labs, Consulting Services
Advanced technical
Talk
50 minutes
Session Schedule