This paper aims to explore the feasibility of neural architecture search (NAS) given only a pre-trained model without using any original training data. This is an important circumstance for privacy protection, bias avoidance, etc., in real-world scenarios. To achieve this, we start by synthesizing usable data through recovering the knowledge from a pre-trained deep neural network. Then we use the synthesized data and their predicted soft-labels to guide neural architecture search. We identify that the NAS task requires the synthesized data (we target at image domain here) with enough semantics, diversity, and a minimal domain gap from the natural images. For semantics, we propose recursive label calibration to produce more informative outputs. For diversity, we propose a regional update strategy to generate more diverse and semantically-enriched synthetic data. For minimal domain gap, we use input and feature-level regularization to mimic the original data distribution in latent space. We instantiate our proposed framework with three popular NAS algorithms: DARTS, ProxylessNAS and SPOS. Surprisingly, our results demonstrate that the architectures discovered by searching with our synthetic data achieve accuracy that is comparable to, or even higher than, architectures discovered by searching from the original ones, for the first time, deriving the conclusion that NAS can be done effectively with no need of access to the original or called natural data if the synthesis method is well designed.
翻译:本文旨在探索神经结构搜索(NAS)的可行性, 仅以事先经过训练的模型为条件, 而不使用任何原始的培训数据。 这是保护隐私、避免偏见等的重要条件, 在现实世界的情景中。 为了实现这一点, 我们首先通过从经过训练的深层神经网络中恢复知识, 合成可用数据。 然后我们使用综合数据及其预测的软标签来引导神经结构搜索。 我们确认NAS任务需要综合数据( 我们在这里图像域的目标), 并有足够的语义、 多样性和自然图像最小域间距。 对于语义学, 我们提议循环标签校准, 以产生更多信息输出。 对于多样性, 我们提出区域更新战略, 以生成更多样化和精密的精密合成数据。 对于最小的域隔间距, 我们使用输入和地平调度来模拟原始数据在潜空空间的分布。 我们以三种受欢迎的NAS 算法即 DARTS、 ProxlesnNAS 和 SPOS 最小域间距差 。 。 令人惊讶的是, 我们建议循环的标签校准校准, 我们的校准的校准校准, 我们的建筑的建筑的校准, 被发现的原结构, 的校准的原结构是无法通过合成的合成的合成的校准, 所发现的校准的校准的校准的校准的校准的校准的校准的校准, 。