Machine Learning. How to Build Reliable Artificial Intelligence Models

This practical guide to creating robust, secure, and interpretable ML systems examines key aspects of developing reliable models: from identifying vulnerabilities and biases to evaluating algorithm transparency, protecting against attacks, and managing liabilities in ML projects.
The book helps understand how modern approaches to fairness, interpretability, and security work, and demonstrates how to apply them in real-world conditions — where models encounter changing environments, noisy data, and human use cases.