During the speech, Professor Wanlei Zhou introduced the background and current development of the concept of Machine Unlearning and proposed some methods for forced forgetting. He pointed out that machine learning has made significant progress in the past decade and has been widely used in application scenarios such as loan applications, facial recognition, recidivism prediction, and large-scale language models, which have had a positive impact on our daily lives. However, due to privacy concerns, security concerns, availability considerations, legal regulations, and other reasons, "machine forgetting" has emerged. 

Professor Wanlei Zhou discussed the operational mechanism behind machine forgetting, as well as its differences from concepts such as catastrophic forgetting, data masking, differential privacy, and online learning. He proposed two methods for forced forgetting - data reorganization and model manipulation - and two methods for verifying whether machine forgetting has been successfully implemented - empirical evaluation and theoretical calculation. Finally, Professor Wanlei Zhou introduced the application scenarios of machine forgetting through practical cases and pointed out the challenges that machine forgetting will face in the future. 

In addition, Professor Wanlei Zhou, Professor Tianqing Zhu, and their research team conducted research on the attack behavior of Graph Neural Networks, proposed a defense method against link stealing attacks, and demonstrated its effectiveness. Their research findings were compiled into a paper called Defense Against Graph Injection Attack in Graph Neural Networks, which was presented by Professor Zhou Wanlei during the conference and received a warm response from the audience. At the same time, they were awarded the Outstanding Paper Award by the conference.