This course will cover the fundamental concepts related to machine learning for cyber security. Topics include:
Prerequisites: Programming skills up to data structures and knowledge of statistics will be useful. No prior experience with machine learning is required.
8:30am-5:00pm
Mon, Tue, Wed, Thu, Fri (End 2:00 PM)
143 Powers
Deep Learning Algorithms
By Ricardo Calix
Ricardo A. Calix, Ph.D.
Purdue University Northwest
rcalix@pnw.edu
Tuesday and Thursday (2-4 PM)
My office is at 241 Anderson. The course will take place in Powers 143
Environment:
AWS
Example problems will be provided as required.
We will use the following software:
Mon | Tue | Wed | Thu | Fri |
---|---|---|---|---|
8:30 AM Lec: ML for Cyber Intro (slides) (video) |
8:30 AM Lec: Performance Metrics (slides) (video) |
8:30 AM Lab: Tensorflow Linear Regression (recap...)(code) |
8:30 AM Multiclass Cross Entropy link Lec: Neural Nets and Deep Neural Nets (slides) (video) |
8:30 AM Lec/Lab: Unsupervised Learning sklearn (video) |
9:00 AM Lec: ML for Cyber Intro (slides) (video) |
9:00 AM Lec: KNN (slides) (video) |
9:00 AM Lec: Deep Learning intro (slides) (video) |
9:00 AM Keras (video) |
9:00 AM Lec/Lab: KMeans (code) and SVD sklearn KMeans with Tensorflow |
10:00 AM break |
10:00 AM break |
10:00 AM break |
10:00 AM break |
10:00 AM break |
10:15 AM Lec: Datasets and Features (slides) (video) |
10:15 AM Naive bayes (slides) (video) Anomaly Detection (slides) (video) |
10:15 AM Lec: Linear Regression (slides) (video) |
10:15 AM Lab: KDD GPU Tensorflow (KDD GPU Lab Doc) (KDD Lab Doc) (video) |
10:15 AM Lec: Deep Reinforcement Learning (slides) (video) (book chapter) |
11:00 AM Lab: WEKA (Iris Lab Doc) (slides) (video) |
11:00 AM Lec, Lab: Sklearn (slides) (video) |
11:00 AM Lec: Linear Regression to Logistic Regression (slides) (video) |
11:00 AM Challenge: Truck Failures |
11:00 AM Lab: Deep Reinforcement Learning (code) (video) |
11:45 AM Short Quiz |
11:45 AM Short Quiz |
11:45 AM Short Quiz (link) |
11:45 AM Short Quiz |
11:45 AM Short Quiz |
12:00 PM lunch |
12:00 PM lunch |
12:00 PM lunch |
12:00 PM lunch |
12:00 PM lunch |
1:00 PM lunch |
1:00 PM lunch |
1:00 PM lunch |
1:00 PM lunch |
1:00 PM |
1:30 PM Lec: IOT Problem Intro (slides) (video) |
1:30 PM Lab: Malware WEKA (Lab Doc) (video) |
1:30 PM Lab: Logistic Regression Tensorflow (code) |
1:30 PM Lab: IOT Tensorflow (Lab Doc) |
1:30 PM Wrap up and Exit Survey |
2:00 PM Lab: IOT Feature extraction (Lab Doc) (video) Lab: IOT WEKA (Lab Doc) (video) |
2:00 PM Lab: IOT Device detection Sklearn (Lab Doc) Lab: Malware sklearn (Lab Doc) (video) |
2:00 PM Lec: Logistic Regression (slides) (video) |
2:00 PM Lab: Malware Tensorflow (Lab Doc) (video) |
2:00 PM |
3:00 PM break |
3:00 PM break |
3:00 PM break |
3:00 PM break |
3:00 PM |
3:15 PM Lec: Malware Feature extraction intro (slides) (video) |
3:15 PM Lec: Data for Tensorflow (slides) (video 1) (video 2) |
3:15 PM Lec: Logistic Regression (slides) (video) |
3:15 PM Lab: Solve Phishing Lab (Lab Doc) (video) |
3:15 PM |
4:00 PM Lab: Malware feature extraction (Lab Doc) (slides) (video) |
4:00 PM Lab: Tensorflow Linear Regression (run and read the code)(code) |
4:00 PM Lab: KDD small (KDD-small Lab Doc) (video) |
4:00 PM Challenge: Credit card fraud (solve this with all your tools and try to achieve best performance metrics) (dataset) (info) |
4:00 PM |
4:30 PM Problem Quiz (Link) |
4:30 PM Problem Quiz |
4:30 PM Problem Quiz |
4:30 PM Challenge: Build best classifier to improve metrics of minority class train test |
4:30 PM |
5:00 PM |
5:00 PM |
5:00 PM |
5:00 PM |
5:00 PM |