Information

• Dates: 27th and 28th of November, 2021

• Location: Online via Zoom.

• Registration: Please register here. (it is free.)

We thank all for participating in the workshop!

• Some of the slides are available for downloading. Please check the program below.

Confirmed Speakers

• Jacek Gondzio (University of Edinburgh)

• Florian Jarre (HHU Düsseldorf)

• Tomonari Kitahara (Kyushu University)

• Masakazu Kojima (Chuo University)

• Bruno F. Lourenço (ISM)

• Tom Luo (CUHK-Shenzhen)

• Shinji Mizuno (Tokyo Institute of Technology)

• Renato Monteiro (Georgia Tech)

• Keiichi Morikuni (University of Tsukuba)

• Masakazu Muramatsu (UEC)

• Gábor Pataki (UNC at Chapel Hill)

• Kunio Tanabe (ISM)

• Mirai Tanaka (ISM)

• Tamás Terlaky (Lehigh University)

• Takashi Tsuchiya (GRIPS)

• Levent Tunçel (University of Waterloo)

• László Végh (London School of Economics)

• Yinyu Ye (Stanford University)

• Akiko Yoshise (University of Tsukuba)

• Yin Zhang (CUHK-Shenzhen)

Program

All times below are Tokyo Time (UTC+09)

For time zone conversions, see this link:

Saturday, 27th of November

9:55: Opening Remarks

Session 1 - 10:00~

• 10:00: Renato Monteiro. Complexity of a Dampened Proximal ADMM for Linearly-constrained Nonseparable Nonconvex Composite Optimization

• 10:40: Tom Luo. Learning to maximize a convex quadratic function with application to intelligent reflection surface for wireless communication

• 11:20: Yinyu Ye. Recent Developments of Online Linear Programming

Session 2 - 13:50~

• 13:50: Mirai Tanaka: A gradient method for multilevel optimization [Slides]

• 14:30: Kunio Tanabe: My Personal Experience in Modelling and Algorithm of Optimization for the Learning Machines PLRM/dPLRM

Session 3 - 15:30~

• 15:30: Bruno F. Lourenço: Completely solving general SDPs [Slides]

• 16:10: Masakazu Kojima: A geometric analysis of strong duality in conic optimization [Slides]

Session 4 - 17:10~

• 17:10: Keiichi Morikuni: Implementation of interior-point methods for LP using Krylov methods preconditioned by inner iterations [Slides]

• 17:50: Jacek Gondzio: Krylov-based Solvers for Interior Point Methods [Slides]

Sunday, 28th of November

Session 1 - 10:00~

• 10:00: Levent Tunçel: A biased view of primal-dual interior-point algorithms for convex optimization and some recent developments

• 10:40: Gábor Pataki: How do exponential size solutions arise in semidefinite programming?

• 11:20: Tamás Terlaky: Quantum Interior Point Methods for LO and Conic Linear Optimization

Session 2 - 13:50~

• 13:50: Tomonari Kitahara: A bound for the number of iterations by the simplex method with the steepest-edge rule

• 14:30: Akiko Yoshise: Evaluating approximations of the semidefinite cone with trace normalized distance [Slides]

Session 3 - 15:30~

• 15:30: Yin Zhang : Distributed (Sparse) PCA: Seeking Consensus on a Subspace

• 16:10: Shinji Mizuno: Approximation algorithms for a single-machine scheduling problem with a non-renewable resource

Session 4 - 17:00~

• 17:00: Florian Jarre: On a very simple analysis of higher order liftings for binary problems

• 17:40: László Végh: On a scaling-invariant layered least squares interior point method

Session 5 - 18:40~

• 18:40: Masakazu Muramatsu: A simple SDP example having different duality gaps during FRA [Slides]

• 19:20: Takashi Tsuchiya: A new look at duality theory for singular SDPs and its implication to convergence of infeasible interior-point algorithms [Slides]

Organizers

• Bruno F. Lourenço (Institute of Statistical Mathematics)

• Keiichi Morikuni (University of Tsukuba)

• Mirai Tanaka (Institute of Statistical Mathematics)