confopt统一了梯度搜索NAS的开发与评估，解决方法分散和评测失真问题。
confopt: A Library for Implementation and Evaluation of Gradient-based One-Shot NAS Methods
- 提供可扩展接口，支持灵活构建新搜索空间和分解优化器组件
- 设计新基准与评估协议，发现现有方法评测存在严重偏差
- 适合研究者快速实现、对比和复现梯度型单次NAS方法
基于梯度的单次神经架构搜索（NAS）显著降低了在离散设计选择（如模型内操作选取）下探索架构空间的成本。然而，该领域面临两大挑战：其一，现有评估高度依赖DARTS基准，导致性能提升趋于饱和，多数改进落入噪声范围；其二，相关方法实现分散于多个独立仓库，难以公平比较与复现。本文提出可配置优化器（confopt），一个可扩展的库，旨在简化梯度型单次NAS方法的开发与评估。confopt 提供轻量级API，便于用户集成新搜索空间，并支持将NAS优化器解耦为核心组件。我们基于此框架构建了一套新的基于DARTS的基准，并结合新型评估协议，揭示了当前梯度型单次NAS方法评估方式中的关键缺陷。代码开源地址：https://github.com/automl/ConfigurableOptimizer。
原文摘要 · Abstract (English)
Gradient-based one-shot neural architecture search (NAS) has significantly reduced the cost of exploring architectural spaces with discrete design choices, such as selecting operations within a model. However, the field faces two major challenges. First, evaluations of gradient-based NAS methods heavily rely on the DARTS benchmark, despite the existence of other available benchmarks. This overreliance has led to saturation, with reported improvements often falling within the margin of noise. Second, implementations of gradient-based one-shot NAS methods are fragmented across disparate repositories, complicating fair and reproducible comparisons and further development. In this paper, we introduce Configurable Optimizer (confopt), an extensible library designed to streamline the development and evaluation of gradient-based one-shot NAS methods. Confopt provides a minimal API that makes it easy for users to integrate new search spaces, while also supporting the decomposition of NAS optimizers into their core components. We use this framework to create a suite of new DARTS-based benchmarks, and combine them with a novel evaluation protocol to reveal a critical flaw in how gradient-based one-shot NAS methods are currently assessed. The code can be found at https://github.com/automl/ConfigurableOptimizer.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。