Machine Unlearning

This book explores one of the most critical and emerging fields in artificial intelligence (AI): machine unlearning. As data privacy concerns grow and regulations like GDPR (General Data Protection Regulation) demand compliance, this book provides a comprehensive guide to selectively removing learned information from machine learning models without sacrificing performance or requiring complete retraining. Covering foundational principles, advanced algorithms, benchmarking tools, and real-world case studies in healthcare, finance, and social media, the book bridges the gap between theory and practice. It also addresses ethical, legal, and societal implications, offering insights into creating trustworthy AI systems. This book is an essential resource for understanding and implementing machine unlearning in the era of responsible AI.

Bitan Misra is currently working as an assistant professor in the Department of Computer Science and Engineering, Techno International New Town, Kolkata, India. She received a B. Tech. and M. Tech. dual degree in electronics and telecommunication engineering from KIIT University, Bhubaneswar, India, in 2018. She received a Ph.D. in 2022 from the National Institute of Technology, Durgapur, India. She has published almost 40 research papers in various international journals and conferences and is the author of five books. Her main research interests include optimization techniques, deep learn­ing, evolutionary algorithms, and soft computing techniques. She has worked as a reviewer in several national and international journals and conferences. She is an associate editor of International Journal of Ambient Computing and Intelligence. She is a member of IEEE and Internet Society. Sayan Chakraborty (senior member, IEEE) is cur­rently working as an assistant professor in the Department of Computer Science and Technology at JIS College of Engineering, Kalyani, West Bengal, India. He completed a Ph.D. in image registration from Sikkim Manipal University in 2023. He obtained an M. Tech. degree in computer sci­ence and engineering from JIS College of Engineering and obtained a B. Tech. degree from the same institution. He has over 12 years of academic experience. His research interests include digital image processing, nature-inspired algorithms, and machine learning. He has published 92 research papers in international journals, book chapters, and conference proceedings on topics including optimization, artificial intelligence, pattern recognition, and digital image processing. Dr. Chakraborty has authored and edited four books published by Springer, Academia Press, and CRC Press. He is currently serving as an associate editor of the International Journal of Ambient Computing and Intelligence and is an Editorial Board member of the International Journal of Rough Sets and Data Analysis. Nilanjan Dey (senior member, IEEE) received B. Tech. and M. Tech. degrees in information technology from West Bengal Board of Technical University and a Ph.D. in electronics and telecommunication engineering from Jadavpur University, Kolkata, India, in 2005, 2011, and 2015. Currently, he is a professor with Techno International New Town, Kolkata, and a visiting fellow of the University of Reading, UK. He is the Editor-in-Chief of the International Journal of Ambient Computing and Intelligence, Associate Editor of IEEE Transactions on Technology and Society, and series Co-Editor of Springer Tracts in Nature-Inspired Computing and Data-Intensive Research and Advances in Ubiquitous Sensing Applications for Healthcare. He is on the Editorial Board of IEEE Data Descriptions. He is a fellow of Institution of Electronics and Telecommunication Engineers and member of Institution of Engineers and Internet Society.

Preface 1. Introduction to Machine Unlearning 2. Technological Approaches to Unlearning 3. Machine Unlearning in Generative AI and LLMs 4. Benchmark Datasets and Experimental Frameworks 5. Case Studies in Machine Unlearning 6. Data Privacy Ethical Implications 7. Challenges in Applying RTBF to AI Systems 8. Conclusions and Future Research Directions