My main research interest is the application of statistics and numerical optimization technology in big data and machine learning, including Bayesian statistics, Bayesian non parameter, asymptotic statistics, concentration equality and other statistical technologies, including the optimization methods such as stochastic gradient descent, coherent gradient method, EM method, etc. At the same time, I am also engaged in the development of high-performance statistics, numerical optimization tools and distributed big data statistics middleware based on SIMD, OpenMP and other parallel technologies. At present, under the leadership of Professor Joshua Zhexue Huang, I am the person in charge of the "i-nice algorithm research and development center". I am engaged in the source code, system development and algorithm design of i-nice (identifying the number of clusters and initial cluster directors) model, gamma texture, C numerical optimization tool library, etc.
Jianfei Yin, Ruili Wang, Yizhe Bai, Shunda Ju, Joshua Zhexue Huang. An Online Portfolio Selection based on Learning Wealth Flow Matrices, ACM TKDD, 2020, under 3rd viewing.
Yizhe Bai, Jianfei Yin, Shunda Ju, Zhao Chen, and Joshua Zhexue Huang. Long and Short Term Risk Control for Online Portfolio Selection. KSEM2020, accepted.
J. Yin, R. Wang, S. Ju, Y. Bai and J. Z. Huang, "An Asymptotic Statistical Learning Algorithm for Prediction of Key Trading Events," in IEEE Intelligent Systems, vol. 35, no. 2, pp. 25-35, 1 March-April 2020, doi: 10.1109/MIS.2020.2977590.
Timur Valiullin, Joshua Zhexue Huang, Jianfei Yin, Dingming Wu: A New Approach for Approximately Mining Frequent Itemsets. DAMDID/RCDL 2019: 46-58，Kazan Federal University, Kazan, Russia, 15-18 Oct 2019
Mengtao Lu and Jianfei Yin*: A Feature Metric Algorithm Combining the Wasserstein Distance and Mutual Information. IEEE International Conference on Progress in Informatics and Computing (PIC) 2018: 154-157, Suzhou, China, 14-16 Dec 2018
Research and development of self-controllable software i-nice for unsupervised learning of complex big data, Shenzhen Natural Science Foundation, Project Leader
Huawei Electronic Assembly Machining Big Data Mining Capability Exploration Project, 2017-2018.5，Project Leader
Research on Approximate Computing Theory and Algorithms for Distributed Large Data Based on Random Sample Partition，Application Project of National Natural Science Foundation of China，Major participants
More info is here http://bigdata.szu.edu.cn/teams?tid=1&id=14