null
 

Frontiers of Cyber Security Workshop

Thursday, November 8, 2018
10:00 am - 4:30 pm

Location:

NYU Tandon School of Engineering
1 MetroTech Center, 19th floor, room 1930/31
Brooklyn, NY 11201

Please check in at the 6 MetroTech Center lobby first to receive your badge.

Cost: 
Workshop cost includes a seat to both workshop sessions (described below), coffee/tea, lunch, and a ticket to the CSAW'18 Welcome Reception. Attendees will receive an official certificate of completion. 

$700.00 

AGENDA 

10:00 - 10:30 am     

Workshop Check-in, Lobby of 6 MetroTech Center

10:30 am

Welcome Address, 1 MetroTech Center, 19th floor

10:45 am - 1:00 pm     

Session I: Data Analytics for Security

Presented by: Damon McCoy

 Security Analytics is a collection of scalable data analysis methods that are able to combine and sift through data from internal (i.e., logs, network traces) and external (i.e., threat intelligence) source. When these methods are successful, they render massive, unwieldy quantities of data into something that is useful and actionable to detect and mitigate attacks. However, the data and methods are often complex to understand and nuanced in the pitfalls of security analytics.

In my talk, I will discuss the data sources and how they might be combined using "big data" tools. When talking about the data I will mention some of the common limitations that we have found. Next, I will talk about how to analyze this data using standard statistical methods and more advanced machine learning techniques. This will focus on the challenge of creating scalable and sustainable analysis methods in the face of a persistent and adaptable adversary. Finally, I will talk about some of the cutting edge methods that hold the possibility of improving the usefulness of security analytics to produce actionable information to detect and mitigate sophisticated attackers.

1:00 pm - 2:00 pm        

Lunch + Networking Mixer

2:00 pm - 4:30 pm        

Session II: Security of AI Systems

Presented by Siddharth GargBrendan Dolan-Gavitt

Machine learning (ML) systems are being pervasively deployed in wide range of application domains from automated driving to image and speech recognition, security analytics (see previous talk) and medical informatics to name a few. However, recent work has shown that ML systems are vulnerable to attack; these include attacks can compromise training data or the training algorithm, introduce vulnerabilities in supply chain used to share and exchange ML models, or engineer targeted modifications to inputs that cause ML systems to misbehave. This talk will provide a broad overview of the ML attack surface and discuss best practices for the safe and secure deployment of ML systems. Case studies of an ML based spam filtering application and a simple autonomous navigation application will be used to illustrate specific vulnerabilities.

Related article: https://www.wired.com/story/machine-learning-backdoors/

5:00 - 7:30 pm  

CSAW'18 Welcome Reception, Keynote Address & Cyber Journalism Award

5 MetroTech Center, Pfizer Auditorium

Attendees of this workshop are invited to attend the CSAW'18 Welcome Reception, followed by the Keynote and the Cyber Journalism Award presentations. 

Register

Register to attend the CSAW'18 Frontiers of
Cyber Security Workshop