Completed
Introduction
Class Central Classrooms beta
YouTube playlists curated by Class Central.
Classroom Contents
Owned By Statistics - How Kubeflow & MLOps Can Help Secure Your ML Workloads
Automatically move to the next video in the Classroom when playback concludes
- 1 Introduction
- 2 Machine Learning at Microsoft
- 3 ML in every product at Microsoft
- 4 ML in the average enterprise
- 5 Data scientist
- 6 Building a model
- 7 Rolling it out
- 8 Security
- 9 Three types of attacks
- 10 Advanced models
- 11 Snow detection
- 12 Stop sign detection
- 13 Face recognition
- 14 Defend against adversaries
- 15 Build an MLOps pipeline
- 16 Modular components
- 17 Pipeline example
- 18 Another attack vector
- 19 Malicious users
- 20 Two types of attacks
- 21 Distillation attack
- 22 Accuracy
- 23 GoogleBERT
- 24 Continuous Improvement
- 25 Build Efficient Pipelines
- 26 Take Your Models
- 27 Hidden Data
- 28 Recommendations
- 29 Network Graph
- 30 Map Leakage
- 31 Example
- 32 How to prevent this
- 33 Injections
- 34 Leaks
- 35 Summary
- 36 The Reality
- 37 You will be attacked
- 38 Conclusion
- 39 Questions
- 40 Reprocessing ML Pipeline Predictions
- 41 MLOps vs Continuous Machine Learning
- 42 Regulation of ML
- 43 Mitigating Leaky Data
