Research map / Learning theory and algorithms / Combinatorial and search optimization
Learning the Pareto Front with Hypernetworks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 a preference-conditioned PHN instead of training preference-specific models | Multi-Fashion+MNIST · Multi-Fashion+MNIST | HV: 2.76 -> 2.78 | Helped |
| Use a preference-conditioned PHN instead of training preference-specific models | PHN-EPO · Multi-Fashion | HV: 2.16 -> 2.19 | Helped |
| Use a preference-conditioned PHN instead of training preference-specific models | PHN-LS · Multi-MNIST | HV: 2.88 -> 2.90 | Helped |
| Use a preference-conditioned PHN instead of preference-specific models | Fairness datasets · Adult, Default, and Bank | HV: Adult: 0.628 -> 0.658; Default: 0.548 -> 0.551; Bank: 0.713 -> 0.748 | Helped |
| Use PHN-EPO instead of preference-specific CPMTL | NYUv2 · NYUv2 | HV: 3.570 -> 3.589 | Helped |
| Use PHN-EPO instead of preference-specific LS | SARCOS · SARCOS | HV: 0.688 -> 0.728 | Helped |
| Use PHN-EPO instead of PHN-LS | PHN · Multi-Fashion+MNIST | Uniformity: 0.894 -> 0.952 | Helped |
| Use PHN-EPO instead of PHN-LS | PHN · SARCOS | HV: 0.693 -> 0.728 | Helped |
| Use PHN-EPO instead of PHN-LS | PHN · Multi-MNIST | HV: 2.90 -> 2.78 | Hurt |
| Use PHN-EPO instead of PHN-LS | PHN · Adult | HV: 0.658 -> 0.648 | Hurt |
| Evaluate PHN on unseen preference rays rather than train rays | PHN · SARCOS | HV: 0.7337 -> 0.7336 | No clear effect |
| Vary the Dirichlet sampling parameter alpha (ablation) | PHN-EPO · Multi-Fashion+MNIST | HV | No clear effect |
Closest papers by meaning
- IOP: An Idempotent-Like Optimization Method on the Pareto Front of Hypernetwork (AAAI 2025 · similarity 0.93)
- Voronoi-grid-based Pareto Front Learning and Its Application to Collaborative Federated Learning (ICML 2025 · similarity 0.91)
- Interactive Hyperparameter Optimization in Multi-Objective Problems via Preference Learning (AAAI 2024 · similarity 0.89)
- Improving Pareto Set Learning for Expensive Multi-objective Optimization via Stein Variational Hypernetworks (AAAI 2025 · similarity 0.89)
- Pareto Low-Rank Adapters: Efficient Multi-Task Learning with Preferences (ICLR 2025 · similarity 0.88)
- How to Find the Exact Pareto Front for Multi-Objective MDPs? (ICLR 2025 · similarity 0.87)
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.