Every tens of years the fundemental philosophy that drives the reserach of machine intelligence shifts between empirism and rationalism. Following the recent success of machine learning and especially deep learning on many tasks such as speech recogonition, computer vision, natural languange processing, information retreival and recommeneder systems, the modeling and utilization of knowledge graph is no longer restricted to hard-rule inference approaches. Indeed, the integrateion of structured knolwedge and embedding learning has provided us with many new techniques to represent, inference, and make predictions based on structured knowledge graphs, and equipping the systems with knowledge again is important to the future of many intelligent sysetms, including but not limited to search engines, recommender systems, online adversiting systems, and many others, which can help to improve both the system performance and the explainability in the future. This project devise efforts to develop new models for learning knowledge graph embeddings, new approaches to reason over knowledge graphs, and new applications based on the learned knowledge graph embeddings.
 Qingyao Ai, Vahid Azizi, Xu Chen, Yongfeng Zhang. Learning Heterogeneous Knowledge Base Embeddings for Explainable Recommendation. Algorithms. 2018, 11(9).
 Liu Yang, Minghui Qiu, Chen Qu, Jiafeng Guo, Yongfeng Zhang, Bruce Croft, Jun Huang, Haiqing Chen. Response Ranking with Deep Matching Networks and External Knowledge in Information-seeking Conversation Systems. In Proceedings of the 41th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2018), July 8 – 12, 2018, Ann Arbor, Michigan, USA
 Yongfeng Zhang, Qingyao Ai, Xu Chen, and W. Bruce Croft. Joint Representation Learning for Top-N Recommendation with Heterogenous Information Sources. In Proceedings of the 26th ACM International Conference on Information and Knowledge Management (CIKM 2017), November 6 – 10, 2017, Singapore.