Completed
Influence Analysis by Gradient Computation
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Neutaint- Efficient Dynamic Taint Analysis with Neural Networks
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Dynamic taint analysis (DTA)
- 3 Limitations of traditional rule-based DTA
- 4 A motivating example: taint propagation
- 5 A motivating example: Neural program embedding
- 6 Neutaint: a new way to track taint information
- 7 Data collection
- 8 Influence Analysis by Gradient Computation
- 9 Evaluation
- 10 Hot byte accuracy, runtime
- 11 Case study: overtaint
- 12 Case study: runtime overhead
- 13 Exploit analysis
- 14 Fuzzing
- 15 Performance on different ML models
- 16 Effect of training data on information loss
