Research map / Optimization / Optimizers
Multi-Objective Meta Learning
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 MOML instead of MAML | few-shot learning · CUB 1-shot | PGD accuracy (%): 4.08 -> 25.67 | Helped |
| Use MOML instead of MAML | few-shot learning · CUB 1-shot | clean accuracy (%): 54.41 -> 47.74 | Hurt |
| Use MOML instead of MAML | few-shot learning · CUB 5-shot | PGD accuracy (%): 8.95 -> 43.41 | Helped |
| Use MOML instead of MAML | few-shot learning · mini-ImageNet 5-shot | PGD accuracy (%): 4.70 -> 38.95 | Helped |
| Use MOML instead of MAML | few-shot learning · mini-ImageNet 1-shot | clean accuracy (%): 46.58 -> 40.03 | Hurt |
| Use MOML-V1 instead of DARTS-C26 at similar size | NAS · CIFAR-10 | PGD accuracy (%): 28.45 -> 42.66 | Helped |
| Use MOML-V1 instead of DARTS-C26 at similar size | NAS · CIFAR-10 | clean accuracy (%): 96.91 -> 96.48 | Hurt |
| Use MOML-V3 instead of DARTS-C34 at similar size | NAS · CIFAR-10 | PGD accuracy (%): 30.31 -> 35.22 | Helped |
| Replace MGDA with equal-weight combination in upper level (w/o MGDA) (ablation) | MOML · CIFAR-10 NAS V1 | PGD accuracy (%): 42.66 -> 40.20 | Hurt |
| Replace MGDA with equal-weight combination (ablation) | MOML · CIFAR-10 NAS V2 | PGD accuracy (%): 36.15 -> 31.44 | Hurt |
| Replace MGDA with equal-weight combination (ablation) | MOML · CIFAR-10 NAS V3 | PGD accuracy (%): 35.22 -> 35.36 | No clear effect |
| Replace MGDA with combined single objective (ablation) | MOML · Office-31 SSDA average | accuracy (%): 90.02 -> 89.85 | Hurt |
Closest papers by meaning
- Enhancing Meta Learning via Multi-Objective Soft Improvement Functions (ICLR 2023 · similarity 0.91)
- A Nested Bi-level Optimization Framework for Robust Few Shot Learning (AAAI 2022 · similarity 0.87)
- Submodular Meta-Learning (NeurIPS 2020 · similarity 0.87)
- Memory-Reduced Meta-Learning with Guaranteed Convergence (AAAI 2025 · similarity 0.86)
- SMOG: Scalable Meta-Learning for Multi-Objective Bayesian Optimization (NeurIPS 2026 · similarity 0.85)
- Learning to Learn with Contrastive Meta-Objective (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.