Research at FAST

105 Department of Applied Mathematics Department of Applied Mathematics Qualification BSc (CQU) MSc (CAS) PhD (UNSW) ORCID ID 0000-0002-5583-4032 Representative Publications • Wa n g J . H . , H u Y. H . , Yu . C . K .W. , L i C . a n d Yang X.Q. , Extended Newton methods for multiobjective optimization: majorizing func tion technique and convergence analysis. SIAM J. Optim. 29 ( 2019 ), no. 3, 2388–2421 • Li M.H., Meng K.W. and Yang X.Q. On error bound moduli for locally Lipschitz and regular functions, Math. Program. Vol. 171 ( 2018 ) no. 1-2, 463-487 • Hu Y.H., Li C., Meng K.W., Qin J. and Yang X.Q. Group sparse optimization via regularization. Journal of Machine Learning and Research 18 ( 2017 ) Paper No. 30, 52 • Hu Y.H.; Li, C. and Yang, X.Q . On convergence rates of linearized proximal algorithms for convex composite optimization with applications. SIAM J. Optim. 26 ( 2016 ), no. 2, 1207–1235 • Fang Y.P., Meng K.W. and Yang X.Q. Piecewise linear multi-criteria programs: the continuous case and its discontinuous generalization. Operations Research Vol. 60 ( 2012 ) 398–409 • Fang D.H., Li C. and Yang X.Q. Stable and total Fenchel duality for DC optimization problems in locally convex spaces. SIAM J. Optim. Vol. 21 ( 2011 ) 730–760 • Meng K.W. and Yang X.Q. Optimality Conditions via exact penalty functions.SIAM J. Optim. Vol. 20 ( 2010 ) 3208–3231 Awards and Achievements • The President’s Awards, The Hong Kong Polytechnic University, for outstanding performance and achievements in Research and Scholarly Activities 2017/18 • The Faculty Award for Outstanding Performance/Achievement - Research and Scholarly Activities 2016/17 • Conference Plenary Talk, Piecewise Linear Multicriteria Programs, Annual Meeting of Chinese Society of Operations Research, Xuzhou Normal University, 17-20 October 2014 Prof. YANG Xiaoqi Professor Research Overview We have studied the local error bounds modulus for locally Lipschitz and regular functions and have provided a geometric meaning of the upper estimate of the modulus by virtue of the distance from 0 to the outer limiting subdifferential of the support function of the subdifferential. We have investigated an extended Newton method for approaching a Pareto optimum of a multiobjective optimization problem, established quadratic convergence criteria and estimated a radius of convergence ball. Numerical performance of the extended Newton method