Causal AI
Robert Osazuwa Ness
Description An introduction to building AI models that identify and reason about cause-and-effect relationships rather than just statistical correlati...
Building Explainable Machine Learning Systems
Ajay Thampi
Publisher
Simon and Schuster
Published
2024
Duration
8 hr 38 min
ISBN
9781617297649
Genres
Description A technical guide to making machine learning models transparent and accountable. It details the methodologies used to "open the black box" of complex models to improve trust, ensure compliance, and mitigate unintended bias.Key Topics: White-box models (linear regression, decision trees, GAMs), model-agnostic methods (LIME, SHAP, Anchors), saliency mapping, network dissection, and fairness mitigation. About the Technology: Interpretability involves identifying patterns learned by a model to explain its results. This is essential for building GDPR-compliant systems and reducing errors from data leakage or concept drift. About the Book: Using Python and open-source libraries, the book demonstrates how to implement interpretability techniques during the development and testing phases of both transparent and deep learning models. About the Reader: For data scientists and engineers familiar with Python and the basics of machine learning. About the Author: Ajay Thampi is a machine learning engineer specializing in responsible AI and model fairness.