Towards Explainable Machine Learning
summary
Describes using AutoML-built, black-box models at Firefly.ai and how interpretability techniques were applied to explain their predictions. Covers sensitivity analysis, partial dependence plots (PDP) & individual conditional expectation (ICE), and LIME, with discussion of pros/cons, differences and production considerations illustrated by two real-world banking and real-estate use cases.
AI-written summary · about
chapters (9)
- 0:00Start
- 2:08Importance of Interpretability High-risk applications
- 5:09Types of Interpretability
- 6:15Complex models
- 12:22Transformation methods: Permutation
- 15:41Pros and Cons
- 17:41Contract Term - PDP
- 22:57Local Explanations
- 27:44Summary
related talks
46mExplaining a Machine Learning blackbox
17mChallenges Towards Production-Ready Explainable Machine Learning
16mBuilding Trust through Explainable AI
01h 29mNavigating ML Explainability: From Theory to Implementation
24mWhy and How to Build Explainability into your ML Workflow
28mManaging AI Risk: Beyond Explainability
35mWhy Do You Need an Explainable Ai?
45mExplainable Machine Learning Explained