Machine Unlearning: A Survey

Machine learning has attracted widespread attention and evolved into an enabling technology for a wide range of highly successful applications, such as intelligent computer vision, speech recognition, medical diagnosis, more. Yet special need arisen where, due to privacy, usability, and/or the right be forgotten , information about some specific samples needs removed from model, called machine unlearning. This emerging drawn significant interest both academics industry its innovation practicality. At same time, this ambitious problem led numerous research efforts aimed at confronting challenges. To best our knowledge, no study analyzed complex topic or compared feasibility existing unlearning solutions in different kinds scenarios. Accordingly, with survey, we aim capture key concepts techniques. The are classified summarized based on their characteristics within up-to-date comprehensive review each category’s advantages limitations. survey concludes by highlighting outstanding issues techniques, along feasible directions new opportunities.