CA$49 – CA$299

Premium Hands-on Workshop: Machine Learning in Cyber Security

Event Information

Share this event

Date and Time

Refund Policy

Refund Policy

Contact the organizer to request a refund.

Eventbrite's fee is nonrefundable.

Event description
ML for CS

About this Event

Workshop Overview

This workshop is an overlap of the two of the hottest topics of the upcoming decade: Machine Learning and Cybersecurity. Cybersecurity is the ultimate data problem. It has massive, often streaming volumes, usually unlabeled, extremely imbalanced data from heterogeneous sources. From the exploding number of IoT devices joining your network, to the carefully crafted spear-phishing email sent to an executive, to the benign-looking image file being sent by a botnet to its C2C - threats, and data, are everywhere. The challenge is further compounded by the wide variation in network environments, meaning abnormal behaviour on an enterprise network is a Tuesday on a university campus. Moreover, threat actors, like any other criminals, try to hide in plain sight, blurring the lines between the malicious and benign.

In this workshop, participants will learn hands on approaches to overcoming obstacles in cyber analytics including absence of reliable labels and extremely large class imbalance, as well as using graph analysis for mining interesting entity relationships, to catch fraudsters who steal money from your bank account.

Important Dates

Early Bird Deadline: January 30, 2020

Refund Deadline: January 30, 2020

Sales end on: February 5, 2020

Please note that this workshop will happen on 3 separate evenings:

Session 1: February 6, 2020

Session 2: February 13, 2020

Session 3: February 20, 2020

"Why should I care about Machine Learning about Cybersecurity?"

If you are working cyber-threat hunting, or pen-testing, or even in a field remotely related to digital or physical security, such as fraud detection, AML, or content moderation, you already have too much data. Here we present some approaches to tackling some of the common problems faced in cybersecurity.

If you are working in ML and are perhaps interested in branching out into cybersecurity, this workshop presents a brief snapshot of the current landscape to which you can bring your expertise.

“How can I attend?”

Join in-person or online

Watch and rewatch the video recordings as often as you need during the 3 weeks of the workshop, and interact with other participants and TAs on our Slack

Need more information?

Send us any questions you might have (events@ai.science), or

Join us for a Q&A session for an opportunity to hear from the instructor and ask them questions about the workshop: https://ai-mlcs-lnl.eventbrite.ca

Target Audience

Data Scientists, Machine Learning Engineers, Other analytics roles, Cyber threat hunters, Pen-testers, fraud analysts, AML analysts

Prerequisites

  • Very comfortable with pandas, sklearn, xgboost, and familiar with PyTorch
  • Environment: google colab
  • Theoretical knowledge assumed: It is expected that the participants are already familiar with the basic statistics, ML, neural network.
  • Basic knowledge of Cybersecurity is nice-to-have

Learning Outcomes

You will learn about different approaches and industry standards for applying ML in Cybersecurity domain. Specifically, you will be learning to:

  • will be able to overcome a lack of labeled data using techniques such as self-supervised learning and domain transfer
  • will be able to avoid pitfalls when evaluating model performance
  • will be able to construct solutions in graph based models

Pre-workshop reading material

  • Adversarial Discriminative Domain Adaptation https://arxiv.org/abs/1702.05464
  • Entity Embedding-Based Anomaly Detection for Heterogeneous Categorical Events https://arxiv.org/abs/1608.07502
  • Evaluating Shallow and Deep Neural Networks for Network Intrusion Detection Systems in Cyber Security
  • Relational Graph Analysis with Real-World Constraints: An Application in IRS Tax Fraud Detection https://www.aaai.org/Papers/Workshops/2005/WS-05-07/WS05-07-006.pdf
  • Optional - Exploring Adversarial Examples in Malware Detection https://arxiv.org/abs/1810.08280

Learning Material

All online and in-person participants will have access to the following learning material:

  • Slides from the sessions
  • Hands on notebooks
  • Video recording of the sessions (you can use the videos to watch the parts that you missed, or re-watch any parts that are still unclear for you; access to videos beyond one week after the workshop is available to be purchased; see tickets >> add-ons)

Instructors

Cathal Smyth (https://www.linkedin.com/in/drcathalsmyth/)

Machine Learning Researcher, RBC

Dr. Smyth is a senior manager of the data analytics group within the Joint Security Operation Centre of RBC. His past experience includes cybersecurity research in the Innovation department of RBC, as well as applied research at Borealis AI. Prior to Joining RBC, Cathal received a post-doctoral fellowship in Big Data at the Fields Institute. Cathal holds a PhD in Physics from the University of Toronto, and has had work presented at various conferences including Blackhat, the Montreal AI Symposium and USENIX Security

Sahar Rahmani (https://www.linkedin.com/in/rahmanisahar/)

Dr. Rahmani is the director of the Analytics team at the Join Security Operation Center at RBC. She manages and leads data scientists in delivering machine learning solutions to detect digital and cyber risks such as fraud, AML, IAM, network security, etc. She makes strategic analytics recommendations based on business and stakeholders needs. Her encouragement and emphasis on innovations in the application of AI/ML in digital risk has resulted in multiple patents, white papers and conference talks for herself and her team members. Sahar holds a PhD in Astrophysics from Western University, where she applied big data analysis and ML techniques to study star formation in nearby galaxies.

Course Modules

The workshop happens on 3 evenings, and covers the following topics: (This is subject to change)

Day 1: Supervised Learning: imbalanced data and domain transfer

  • Introduction and context
  • Improving model performance with imbalanced data
  • Pitfalls in assessing models trained with imbalanced data
  • Adversarial domain adaptation on unlabeled DoS data.
  • (if time) adversarial attacks

Day 2: Self-Supervised learning

  • Introduction to self-supervised learning.
  • Comparison with anomaly detection
  • Noise Contrastive estimation
  • Application (either IDS or botnet detection)

Day 3: Graph-based ML

  • Introduction to graph analysis in cyber security
  • Relational Database vs. graph database
  • Mining entities from graph
  • Anomaly detection using graph analysis

█░ Kick-starter style

Please note that our technical workshops are run kick-starter style in the sense that they will only happen if a certain amount of funds are raised by the 2-week deadline prior to the event (January 23, 2020):

Funds raised so far to enable the workshop:

░░░░░░░░░░ 0%

Discount Codes

There are discount codes available

  • AISC members receive discounts according to their participation credit; refer to the slack channel for more detail

Referral Program

AISC Members, have personalized referral discount codes up to 50% that they can share with their friends and coworkers. Each code can be used 3 times. In order to ensure that people are using the code with your permission they need to enter your email address when registering

Non-members, can enjoy a group discount (around 10% on top of the current early bird discount) by forming groups of 3+ and registering together. So, go talk to your friends or coworkers right now so that you all save 10% more together.

Terms of Service

By signing up for this event you agree to the following terms, privacy policy, and code of conduct: https://ai.science/terms

Share with friends

Date and Time

Refund Policy

Contact the organizer to request a refund.

Eventbrite's fee is nonrefundable.

Save This Event

Event Saved