Explainable Artificial Intelligence (XAI): Why, When, and How?
summary
Covers Explainable AI (XAI) needs across domains like finance, defense, legal and medical, explains trade-offs between model types such as inherently interpretable models (e.g., decision trees) versus using a secondary model to explain a high-accuracy primary model, and presents a framework to analyze choices around explainability, accuracy and development cost along with current challenges and thinking on when and why to require XAI.
AI-written summary · about
related talks
15mIntroduction to Explainable AI
37mTrusted and Responsible AI (Explainability, Adversarial, Bias, Fairness)
47mML: Explainable Artificial Intelligence explained
32mOn the Importance of Explainability
26mUsing Explainable AI to garner trust and overcome adoption inhibitors
30mTrustworthy AI - opening up the black box
28mThe Need for Explainable AI
35mWhy Do You Need an Explainable Ai?