Research map / Learning theory and algorithms / Combinatorial and search optimization
Hypervolume Maximization: A Geometric View of Pareto Set Learning
Xiaoyuan Zhang, Xi Lin, Bo Xue, Yifan Chen, Qingfu Zhang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use hypervolume-based PSL-HV1 instead of EPO-based PSL | PSL-EPO · ZDT1 | HV: 10.82 -> 11.79 | Improved |
| use PSL-HV1 instead of EPO-based PSL, time | PSL-EPO · ZDT1 | time (s): 90.7 -> 31.79 | Improved |
| use PSL-HV2 instead of EPO-based PSL, time | PSL-EPO · MO-LQR3 | time (s): 1149.4 -> 424.51 | Improved |
| use PSL-HV1 instead of linear scalarization | PSL-LS · ZDT2 | HV: 8.75 -> 11.3 | Improved |
| use PSL-HV1 instead of linear scalarization, range | PSL-LS · ZDT2 | Range: 0.0 -> 1.46 | Improved |
| use PSL-HV1 instead of Tchebycheff scalarization | PSL-Tche · Rocket Injector Design | HV: 37.73 -> 40.84 | Improved |
| use PSL-HV1 instead of Tchebycheff scalarization | PSL-Tche · VLMOP1 | HV: 11.83 -> 12.07 | Improved |
| replace PSL-HV1 with PSL-HV2 | PSL-HV1 · ZDT2 | HV: 11.3 -> 11.51 | Improved |
| replace PSL-HV1 with PSL-HV2 | PSL-HV1 · Hatch Cover Design | HV: 12.18 -> 12.15 | No clear effect |
| replace PSL-HV1 with PSL-HV2, sparsity | PSL-HV1 · Rocket Injector Design | Sparsity: 3.31 -> 5.54 | Hurt |
| use hypervolume PSL vs SMS-EMOA time on Rocket Injector | SMS-EMOA · Rocket Injector Design | running time: 6.2m -> <1s | Improved |
| add gradient clipping (norm 4.0) to PSL-HV2 | PSL-HV2 · VLMOP2 | PSL loss | Improved |
Closest papers by meaning
- Improving Pareto Set Learning for Expensive Multi-objective Optimization via Stein Variational Hypernetworks (AAAI 2025 · similarity 0.88)
- Optimal Scalarizations for Sublinear Hypervolume Regret (NeurIPS 2024 · similarity 0.87)
- IOP: An Idempotent-Like Optimization Method on the Pareto Front of Hypernetwork (AAAI 2025 · similarity 0.86)
- Amortized Active Generation of Pareto Sets (NeurIPS 2025 · similarity 0.85)
- Probability Distribution of Hypervolume Improvement in Bi-objective Bayesian Optimization (ICML 2024 · similarity 0.85)
- Preference Optimization on Pareto Sets: On a Theory of Multi-Objective Optimization (NeurIPS 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
