An Overview of Federated Recommendation
COM2 Level 4
Executive Classroom, COM2-04-02
Abstract:
Despite its great progress so far, artificial intelligence (AI) is facing a serious challenge in the availability of high-quality big data. In many real-world applications such as recommender systems, data are decentralized. Efforts to integrate the data are increasingly difficult, due to serious concerns over user privacy and data security. The problem is exacerbated by strict government regulations such as Europe's General Data Privacy Regulations (GDPR). Federated learning is a newly developed technology to bridge data repositories without compromising data security and privacy. In this talk, I will give an overview of federated learning's recent development on recommender systems. In particular, I will introduce a concept of "federated recommendation", its categorization and corresponding pilot studies.
Biodata:
Dr. Zheng is a Deputy General Manager in WeBank, China's very first digital bank. In WeBank, he developed a world-first computational bank-advertising platform, which is in production for inclusive finance. His expertise is on transfer learning, graph learning and federated learning, especially in the context of building large-scale advertising and recommender systems. His work on graph feature engineering is highlighted as IJCAI-ECAI 2018 Early Career Spotlight. His work on deep transfer learning receives the Best Paper Award from ICCSE 2018. He is the Associate Editor of Cognitive Computation. He is a Senior PC in IJCAI 2020, AAAI 2020, and Tutorial Chair in IJCAI 2020. He has published over 70 papers, and held over 40 patents.