Toward Trustworthy AI: Explainability and Robustness
summary
Regulated banking use cases focus on explainability and robustness for credit, fraud, market hedging and trader surveillance; models create financial and legal harm when wrong. The work constrains deep nets into few interpretable components with a projection layer, uses model distillation to replace complex boosters with simple trees or linear pieces, and builds local linear ReLU diagnostics to enumerate per-sample linear equations, compute distance to decision boundaries, and measure how small perturbations flip outcomes.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 3:56Examples of Model Risk
- 7:06Machine Learning Model Race
- 9:12Explainability
- 12:37Selforganizing models
- 17:19Fully connected deep learning
- 20:14Robustness
related talks
28mManaging AI Risk: Beyond Explainability
43mBring in the Lawyers: Explainable AI Driven Decision-making
35mWhy Do You Need an Explainable Ai?
24mWhy and How to Build Explainability into your ML Workflow
16mBuilding Trust through Explainable AI
17mChallenges Towards Production-Ready Explainable Machine Learning
02h 01mFostering Trust in AI-Driven Finance
01h 29mNavigating ML Explainability: From Theory to Implementation