AI Thrust Seminar | Data for AI: Transforming Data Representation Space
In the past years, research has been focusing on optimizing model space in AI. Deep learning models have successfully applied to almost every area. Models trained with millions of parameters and sophisticated neural architectures are now used routinely. It seems models play more role than data. We investigate a question: Can optimizing data space be as powerful as optimizing model space? Relevant representation learning techniques can automatically reconstruct data representation space. But the techniques needs more explainable and traceable explicitness, and flexible optimal. In this talk, we will propose a concept of self-optimizing data geometry. We will introduce explainable and optimal representation space reconstruction from a selection perspective and a generation perspective. Finally, we will discuss our future work.
Dr. Yanjie Fu is an assistant professor in the Department of Computer Science at the University of Central Florida. He received his Ph.D. degree from Rutgers, the State University of New Jersey in 2016, the B.E. degree from University of Science and Technology of China in 2008, and the M.E. degree from Chinese Academy of Sciences in 2011. He has research experience in industry research labs, such as Microsoft Research Asia and IBM Thomas J. Watson Research Center. He has published prolifically in refereed journals and conference proceedings, such as IEEE TKDE, IEEE TMC, ACM TKDD, ACM SIGKDD, AAAI, IJCAI, VLDB, ICDE, WWW, ACM SIGIR, ICLR. His research has been recognized by: 1) two federal faculty awards: US NSF CAREER and NSF CRII awards; 2) four best paper (runner-up, finalist) awards, including IEEE ICDM14 Best Paper Finalist, ACM KDD18 Best Student Paper Finalist, ACM SIGSpatial20 Best Paper Runner-up, IEEE ICDM21 Best Paper Runner-up; 3) five federal sole PI/PI (not coPI) core research program awards; 4) several other university-level awards: University System Research Board Award and University Interdisciplinary Research Award; 5) his graduated Ph.D. students have joined academia (e.g., University of Macau, Portland State University) as tenure-track faculty members. He is broadly interested in data mining, machine learning, and their interdisciplinary applications. His research aims to develop robust machine intelligence with imperfect and complex data by building tools to address framework, algorithmic, data, and computing challenges.