Optimization
Large-scale optimization in finance, machine learning, and operations researchResearch · Hong Kong SAR, China
Optimization research, with photography kept close as a parallel visual notebook.
A quieter front page for Xiwen Wang: research stays unmistakably in front, notes remain one click away, and the gallery works as a curated second register rather than a competing homepage.
02 research notes
06 gallery regions
24 place entries
Numerical computation
Numerical computation: sparse linear systems, matrix reordering, parallel and distributed computingComputational imaging
Image processing: focusing, stabilization, and AI-assisted image editingNotes stay near the front.
A single lead entry gives the homepage a clear second beat, instead of spreading attention across equally weighted modules.
Finance
S&P 500 新纳入成分股公告后路径研究：短窗 alpha 可以交易，长窗 alpha 不宜预设
对 2010–2025 年 S&P 500 新纳入成分股的事件研究。样本显示超额收益主要集中在公告后第一周与生效日前，生效后平均表现更接近回吐，不宜预设长窗持续 alpha。
Travel footprint
A single static map, large enough to read.
This module is now intentionally simple: one locally maintained map image, one stable homepage slot, and no runtime point-projection logic between you and the final result.
24 tracked stops
10 countries / areas
04 macro regions
0 external map deps
Research
Academic identity stays legible from the first scroll.
Three representative papers carry the research profile on the homepage, giving visitors concrete work before they ever need a longer publication list.
Efficient Algorithms for General Isotone Optimization
Algorithmic work on general order-constrained optimization, focused on efficient methods that scale beyond the narrowest special cases.
AAAI · 2022
Efficient and Scalable Parametric High-Order Portfolios Design via the Skew-t Distribution
A portfolio-optimization formulation that brings higher-order moments into a scalable parametric model instead of treating them as a purely theoretical extension.
IEEE Transactions on Signal Processing · 2023
Learning Large-Scale MTP2 Gaussian Graphical Models via Bridge-Block Decomposition
A decomposition-based route for learning large-scale structured graphical models while keeping the computation practical at nontrivial problem sizes.
NeurIPS · 2023
Algorithmic vocabulary
Recurring techniques kept visible without falling back into the old long-CV homepage pattern.
- Active Set Methods
- ADMM
- Block Coordinate Descent
- Fixed Point Acceleration
- Majorization-Minimization
- Projected Gradient Methods
- Quasi-Newton Methods
- Successive Convex Approximation
Based in Hong Kong SAR, China · For research, analysis, or creative collaboration, use Contact.