DEPARTMENT OF APPLIED MATHEMATICS 93 Email houduo.qi@polyu.edu.hk Qualification BSc (Peking University) MSc (Qufu Normal University) PhD (Chinese Academy of Sciences) ORCID ID 0000-0003-3481-4814 Prof. QI Houduo Professor Research Overview I am mainly interested in both theoretical and numerical optimisation with applications to data science. In particular, I am interested in matrix optimisation addressing correlation stress testing in finance, Euclidean distance embedding methods in machine learning and maximum diversification returns in portfolio construction. Furthermore, I specialise in developing Newton's method for sparse optimisation. Representative Publications • Chen Zhao, Naihua Xiu, Houduo Qi, and Ziyan Luo, A Lagrange – Newton algorithm for sparse nonlinear programming, Math. Program., 195 (1-2) (2022), 903–928 • Shenglong Zhou, Lili Pan, Naihua Xiu, and Houduo Qi, Quadratic convergence of smoothing Newton’s method for 0/1-loss optimization, SIAM J. Optim., 31 (4) (2021), 3184– 3211 • Shenglong Zhou, Naihua Xiu, and Houduo Qi, Global and quadratic convergence of Newton hard-thresholding pursuit, J. Mach. Learn. Res., 22 (12) (2021), 1–45 • Chao Ding and Houduo Qi, Convex optimization learning of faithful Euclidean distance representations in nonlinear dimensionality reduction, Math. Program., 164 (1-2) (2017), 341–381 • Houduo Qi, A convex matrix optimization for the additive constant problem in multidimensional scaling with application to locally linear embedding, SIAM J. Optim., 26 (4) (2016), 2564–2590 Achievements • Turing Fellow, The Alan Turing Institute, 2019 • Queen Elizabeth II Fellow, Australian Research Council, 2003
RkJQdWJsaXNoZXIy Mjc5OTU=