PDHCG-II: An Enhanced Version of PDHCG for Large-Scale Convex QP
Hongpei Li, Yicheng Huang, Huikang Liu, Dongdong Ge, Yinyu Ye · 2026
Read PDHCG-II: An Enhanced Version of PDHCG for Large-Scale Convex QP by Hongpei Li, Yicheng Huang, Huikang Liu, Dongdong Ge, Yinyu Ye (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-23967.txt.
Description
Hongpei Li and four co-authors present PDHCG-II, an enhanced first-order solver for large-scale convex quadratic programming. The 2026 paper builds on the recent PDHCG framework, which matters because quadratic programs underpin decision-making and machine learning, yet large instances remain hard to solve efficiently.
The method exploits the quadratic structure of the objective and adds several innovations. One innovation is Halpern-type acceleration. A second is a PID-controlled adaptive update of the primal-dual weight. A third is a refined termination rule for inner subproblems that avoids over-solving. A fourth is an infeasibility detector for ill-posed cases. Tests on standard benchmarks show consistent speedups over PDHCG, by several times. A CUDA-C implementation is released as open source, supporting reproducibility and adoption for large-scale optimization in practice.
Catalog record
| CREATOR | Hongpei Li, Yicheng Huang, Huikang Liu, Dongdong Ge, Yinyu Ye |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-23967