What Testing Strategy for Your Machine Learning Applications?
summary
Oui.sncf describes their testing strategy for production machine-learning applications, addressing non-deterministic outputs when expected results aren’t known. They present a custom testing approach built around three testing layers designed for the uncertainty of ML systems.
AI-written summary · about
chapters (10)
- 0:00ML testing intro
- 6:24Software testing basics
- 8:10Unit tests and TDD
- 13:31Integration tests training/prediction
- 19:51ML performance limits
- 26:59Regression tests on real data
- 31:52Automating ML tests
- 35:39Model degradation monitoring
- 37:45ML testing summary
- 38:06Q&A
related talks
32mTesting Machine Learning Systems: What it is and why you should care
31mWait, You Test Neural Networks Too?
51mHow to effectively test your AI applications?
01h 04mWhat Testers Need from Machine Learning and Artificial Intelligence
53mUnleashing the Black Box: Testing Machine Learning Systems
45mHow to Test Artificial Intelligence and Machine Learning Solutions?
28mML-Driven Test Intelligence at Scale - What Works What Fails and Why It Matters
25mFuture of Software Testing: Machine Learning Assistance