Some recent talks
-
A majorized proximal point dual Newton algorithm for nonconvex statistical optimization problems (The Sixth International Conference on Continuous Optimization, Technical University (TU) of Berlin, Germany, August 3–8, 2019).
-
Matrix Cones and Spectral Operators of Matrices (Advances in the Geometric and Analytic Theory of Convex Cones, Sungkyunkwan University, Korea, May 27–31, 2019).
-
On the Relationships of ADMM and Proximal ALM for Convex Optimization Problems (Institute of Applied Physics and Computational Mathematics, Beijing, April 12, 2019).
-
Sparse semismooth Newton methods and big data composite optimization (New Computing-Driven Opportunities for Optimization, Wuyishan, August 13-17, 2018).
-
On the efficient computation of the projector over the Birkhoff polytope (International Symposium on Mathematical Programming 2018, Bordeaux, July 1-6, 2018).
-
A block symmetric Gauss-Seidel decomposition theorem and its applications in big data nonsmooth optimization (International Workshop on Modern Optimization and Applications, AMSS, Beijing, June 16-18, 2018).
-
On the Equivalence of Inexact Proximal ALM and ADMM for a Class of Convex Composite Programming(DIMACS Workshop on ADMM and Proximal Splitting Methods in Optimization, Rutgers University, June 11-13, 2018).
-
A block symmetric Gauss-Seidel decomposition theorem and its applications in big data nonsmooth optimization (The Hong Kong Mathematical Society Annual General meeting 2018, May 26, 2018).
-
SDPNAL+: A MATLAB software package for large-scale SDPs with a user-friendly interface (SIAM-ALA18, May 2018).
-
Second order sparsity and big data optimization (October 2017).
-
Error bounds and the superlinear convergence rates of the augmented Lagrangian methods (October 2017).
-
Block symmetric Gauss-Seidel iteration and multi-block semidefnite programming (October 2017).
-
A two-phase augmented Lagrangian approach for linear and convex quadratic semidefinite programming problems(December 2016).
-
Linear rate convergence of the ADMM for multi-block convex conic programming (August 2016).
-
An efficient inexact accelerated block coordinate descent method for least squares semidefinite programming (June 2015).
-
Multi-stage convex relaxation approach for low-rank structured PSD matrix recovery (May 2014).