Responsible AI: Implementing Ethical and Unbiased Algorithms

Description

This book is written for software product teams that use AI to add intelligent models to their products or are planning to use it. As AI adoption grows, it is becoming important that all AI driven products can demonstrate they are not introducing any bias to the AI-based decisions they are making, as well as reducing any pre-existing bias or discrimination.  The responsibility to ensure that the AI models are ethical and make responsible decisions does not lie with the data scientists alone. The product owners and the business analysts are as important in ensuring bias-free AI as the data scientists on the team. This book addresses the part that these roles play in building a fair, explainable and accountable model, along with ensuring model and data privacy. Each chapter covers the fundamentals for the topic and then goes deep into the subject matter – providing the details that enable the business analysts and the data scientists to implement these fundamentals.  AI research is one of the most active and growing areas of computer science and statistics. This book includes an overview of the many techniques that draw from the research or are created by combining different research outputs. Some of the techniques from relevant and  popular libraries are covered, but deliberately not drawn very heavily from as they are already well documented, and new research is likely to replace some of it. Foreword Preface Who is this Book for? How to Read this Book How to Access the Code Acknowledgements Contents Chapter 1: Introduction What Is Responsible AI Facets of Responsible AI Fair AI Explainable AI Accountable AI Data and Model Privacy Bibliography Chapter 2: Fairness and Proxy Features Introduction Key Parameters Confusion Matrix Common Accuracy Metrics Fairness and Fairness Metrics Fairness Metrics Equal Opportunity Predictive Equality Equalized Odds Predictive Parity Demographic Parity Average Odds Difference Python Implementation Proxy Features Methods to Detect Proxy Features Linear Regression Variance Inflation Factor (VIF) Linear Association Method Using Variance Cosine Similarity/Distance Method Mutual Information Conclusion Bibliography Chapter 3: Bias in Data Introduction Statistical Parity Difference Disparate Impact When the Y Is Continuous and S Is Binary When the Y Is Binary and S Is Continuous Conclusion Key Takeaways for the Product Owner Key Takeaways for the Business Analysts/SMEs Key Takeaways for the Data Scientists Bibliography Chapter 4: Explainability Introduction Feature Explanation Information Value Plots Partial Dependency Plots Accumulated Local Effects Sensitivity Analysis Model Explanation Split and Compare Quantiles Global Explanation Local Explanation Morris Sensitivity Explainable Models Generalized Additive Models (GAM) Counterfactual Explanation Conclusion Bibliography Chapter 5: Remove Bias from ML Model Introduction Reweighting the Data Calculating Weights Implementing Weights in ML Model Protected Feature: Married Protected Feature: Single Protected Feature: Divorced Protected Feature: Number of Dependants Less than Three Protected Feature: Work Experience Less than 10 Years Calibrating Decision Boundary Composite Feature Additive Counterfactual Fairness High Level Steps for Implementing ACF Model ACF for Classification Problems ACF for Continuous Output Linear Regression Model ACF Model Calculating Unfairness Conclusion Bibliography Chapter 6: Remove Bias from ML Output Introduction Reject Option Classifier Optimizing the ROC Handling Multiple Features in ROC Conclusion Bibliography Chapter 7: Accountability in AI Introduction Data Drift Covariate Drift Jensen-Shannon Distance Wasserstein Distance Stability Index Concept Drift Kolmogorov–Smirnov Test Brier Score Page-Hinkley Test (PHT) Early Drift Detection Method Hierarchical Linear Four Rate (HLFR) Conclusion Bibliography Chapter 8: Data and Model Privacy Introduction Basic Techniques Hashing K-Anonymity, L-Diversity and T-Closeness Differential Privacy Privacy Using Exponential Mechanism Differentially Private ML Algorithms Federated Learning Conclusion Bibliography Chapter 9: Conclusion Responsible AI Lifecycle Responsible AI Canvas AI and Sustainability Need for an AI Regulator Bibliography