Mitigating Bias in Machine Learning 1st Edition

Mitigating Bias in Machine Learning, 1st Edition by Carlotta A. Berry and Brandeis Hill Marshall

Capture confidence in your AI systems with this timely, practice-focused guide to recognizing and reducing bias across the machine learning lifecycle. Designed for data scientists, engineers, policymakers, and advanced students, this book blends clear theory with actionable techniques so you can build fairer, more trustworthy models.

Discover why bias emerges—from data collection and labeling to model selection and deployment—and learn proven strategies for prevention and correction. The authors walk readers through concrete methods: robust data auditing, transparency and interpretability techniques, fairness-aware model training, evaluation using multiple fairness metrics, and practical mitigation workflows adaptable to real-world projects. Case studies and examples show how these approaches apply across industries and regulatory landscapes, making the content globally relevant for practitioners in North America, Europe, Asia, and beyond.

Readable yet rigorous, the book balances statistical foundations with ethical and legal considerations, helping teams align technical decisions with organizational values and compliance requirements. You’ll gain tools to reduce disparate impact, improve stakeholder trust, and communicate risks and solutions to nontechnical audiences.

Whether you’re leading an AI initiative at a startup, updating an enterprise pipeline, or teaching a graduate course on ethical AI, this edition equips you to spot vulnerabilities early and implement scalable remedies. Forward-thinking, practical, and authoritative—this is an essential resource for anyone committed to building machine learning systems that are both effective and equitable.

Order your copy today to make fairness a measurable, repeatable part of your ML practice.

Note: eBooks do not include supplementary materials such as CDs, access codes, etc.