Abstract:Prediction-based approaches are widely used in neural architecture search (NAS), where a predictor estimates the performance of candidate architectures to guide selection. However, existing predictors are typically trained via supervised regression on limited samples, leading to overfitting and poor generalization to unseen architectures. In this work, we propose a fundamentally different formulation that models performance prediction as a conditional function inference problem using a Convolutional Neural Process (ConvNP) with meta-learning capabilities. Instead of fitting a fixed mapping to limited samples, our approach meta-learns to infer performance from partial observations by training with context-target splits across a group of synthesized tasks, explicitly optimizing for generalization under data scarcity and aligning the training procedure with the deployment setting in NAS. We further design simple yet effective meta-features for cell-based architectures and evaluate our method on NAS-Bench-101 and NAS-Bench-201. Extensive experiments show that our approach consistently improves top-K ranking quality and achieves the state-of-the-art architecture selection using limited samples.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.09290 [cs.LG] |
| (or arXiv:2605.09290v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.09290 arXiv-issued DOI via DataCite (pending registration) |
From: Liping Deng [view email]
[v1]
Sun, 10 May 2026 03:34:39 UTC (54 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。