Model interpretability, transparency, and techniques to understand why AI systems make their decisions.
- Coursera: Interpretable Machine Learning - Free to audit course covering major XAI techniques and frameworks.
Intermediate - Google: Explainable AI on Google Cloud - Free tools, documentation, and guides for interpretable AI.
Intermediate - Harvard: Interpretability in Machine Learning - Full course materials on ML interpretability methods.
Advanced
- Interpretable Machine Learning (Christoph Molnar) - The go-to free book on model interpretability — comprehensive and regularly updated.
Intermediate - Fairness and Machine Learning (Barocas, Hardt, Narayanan) - Free textbook on fairness, bias, and accountability in ML.
Intermediate - Explanatory Model Analysis (Biecek & Burzykowski) - Free book exploring model explanations with R examples.
Intermediate
- SHAP (SHapley Additive exPlanations) - Leading explainability library with tutorials and visualizations.
Intermediate - LIME (Local Interpretable Model-agnostic Explanations) - Tools and tutorials for explaining individual predictions.
Intermediate - Captum (PyTorch) - Model interpretability library for PyTorch with comprehensive tutorials.
Intermediate - InterpretML (Microsoft) - Open-source toolkit for training interpretable models and explaining blackbox systems.
Intermediate - Alibi Explain - Open-source library of ML model inspection and interpretation algorithms.
Advanced - What-If Tool (Google) - Interactive visual tool for exploring ML model behavior without code.
Beginner