Research map / Probabilistic and scientific modeling / Bayesian optimization
Interpretable Neural Architecture Search via Bayesian Optimisation with Weisfeiler-Lehman Kernels
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 NAS-BOWL instead of LaNet | DARTS open-domain search on CIFAR-10 · CIFAR-10 | Best Error: $2.61$ -> $2.50$ | Helped |
| Use NAS-BOWL instead of LaNet | DARTS open-domain search · CIFAR-10 | Avg. Error: $2.53_{\pm0.05}$ -> $\mathbf{2.61_{\pm 0.08}}$ | Mixed |
| Use WL graph kernel rather than MLP graph kernel (ablation) | GP regression surrogate · N101 | Spearman rank correlation: 0.458$\pm$0.07 -> 0.862$\pm$0.03 | Helped |
| Use WL graph kernel rather than MLP graph kernel (ablation) | GP regression surrogate · N201 CIFAR-10 | Spearman rank correlation: 0.412$\pm$0.15 -> 0.812$\pm$0.06 | Helped |
| Use WL graph kernel rather than MLP graph kernel (ablation) | GP regression surrogate · N201 CIFAR-100 | Spearman rank correlation: 0.519$\pm$0.14 -> 0.823$\pm$0.03 | Helped |
| Use WL graph kernel rather than MLP graph kernel (ablation) | GP regression surrogate · N201 ImageNet16 | Spearman rank correlation: 0.538$\pm$0.07 -> 0.796$\pm$0.04 | Helped |
| Use WL graph kernel rather than MLP graph kernel (ablation) | GP regression surrogate · Flowers102 | Spearman rank correlation: 0.492$\pm$0.12 -> 0.804$\pm$0.018 | Helped |
| Add an MLP kernel to the WL kernel (ablation) | GP regression surrogate · N101 | Spearman rank correlation: 0.862$\pm$0.03 -> 0.871$\pm$0.02 | Helped |
| Add an MLP kernel to the WL kernel (ablation) | GP regression surrogate · Flowers102 | Spearman rank correlation: 0.804$\pm$0.018 -> 0.813$\pm$0.018 | Helped |
Closest papers by meaning
- Neural Architecture Generator Optimization (NeurIPS 2020 · similarity 0.93)
- BANANAS: Bayesian Optimization with Neural Architectures for Neural Architecture Search (AAAI 2021 · similarity 0.91)
- ProxyBO: Accelerating Neural Architecture Search via Bayesian Optimization with Zero-Cost Proxies (AAAI 2023 · similarity 0.90)
- einspace: Searching for Neural Architectures from Fundamental Operations (NeurIPS 2024 · similarity 0.90)
- Encodings for Prediction-based Neural Architecture Search (ICML 2024 · similarity 0.89)
- Robustifying and Boosting Training-Free Neural Architecture Search (ICLR 2024 · similarity 0.89)
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.