Completed
Readelf model performance over 48h and productization
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Transfer Learning - Repurposing ML Algorithms from Different Domains to Cloud Defense
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Leveraging intelligence across product lines
- 3 Microsoft's cloud security scale - Daily numbers
- 4 Textbook ML development
- 5 Traditional versus Transfer learning
- 6 Why transfer learning
- 7 Detecting malicious network activity in Azure Core Concept: Achieve transfer leaming by grouping similar tasks
- 8 Ensemble Tree Learning applications at Microsoft
- 9 Input data
- 10 Tree Ensembles - Algorithm
- 11 Tree Ensembles - Training
- 12 Tree Ensembles - Testing
- 13 Model performance and productization Model trained at regular intervals
- 14 Bonus Classifier can be used as an effective canary for emerging attacks
- 15 WannaCry Attack Timeline
- 16 Detecting Malicious PowerShell commands Core Concept: Transposing existing security problem into an already solved problem from another domain
- 17 PowerShell command lines - difficult to detect
- 18 Microsoft's Deep Learning toolkit (CNTK) applications
- 19 Deeper learning = representation learning
- 20 Technique overview
- 21 Neural Fuzzing Core Concept: Transposing existing security problem into an already solved problem from another domain
- 22 Seq2Seq Neural Architecture
- 23 Improved fuzzing intuition
- 24 readelf dataset example
- 25 Example readelf 2.28 model
- 26 Analysis by GDB exploitable plugin Target: Linux readelf 2.28
- 27 Readelf model performance over 48h and productization
- 28 Conclusion
- 29 Resources
