Files
myWiki/concepts/zero-cost-proxies.md

30 lines
1.2 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

---
title: "Zero-Cost Proxies (ZCP)"
created: 2026-05-15
updated: 2026-05-15
type: concept
tags: [machine-learning, neural-architecture-search, efficiency]
sources: [raw/papers/zeng-neurida-2025.md]
---
# Zero-Cost Proxies (ZCP)
**Zero-Cost Proxies** 是源自 Neural Architecture Search (NAS) 的技术,在**不进行完整训练的情况下**估计模型在给定任务上的性能。
## 核心思想
在模型初始化阶段通过某些可计算的代用指标(如梯度范数、激活模式、雅可比矩阵特征等)来预测模型的最终性能,成本接近零(无需梯度下降迭代)。
## 在 NeurIDA 中的应用
[[conditional-model-dispatcher|Conditional Model Dispatcher]] 使用 ZCP 对 [[composable-base-model-architecture|基础模型池]] 中每个候选模型进行快速评分实现轻量级的模型选择。ZCP 评分的低成本意味着 Dispatcher 可以在几乎不增加延迟的情况下完成模型选择决策。
## 关键参考文献
- Abdelfattah et al., "Zero-Cost Proxies for Lightweight NAS", ICLR 2021
- Shu et al., "NASI: Label- and Data-agnostic Neural Architecture Search at Initialization", ICLR 2022
## 来源
- [[zeng-neurida-2025|NeurIDA 论文]]