NAS-Navigator: Visual Steering for Explainable One-Shot Deep Neural Network Synthesis

NAS-Navigator: Visual Steering for Explainable One-Shot Deep Neural Network Synthesis
复制标题

NAS-Navigator:用于可解释的一次性深度神经网络合成的视觉引导

DOI:
10.1109/tvcg.2022.3209361
复制
发表时间:
2023
影响因子:
5.2
通讯作者:
Mueller, Klaus
Mueller, Klaus
中科院分区:
计算机科学1区
文献类型:
--
作者:
Tyagi, Anjul;Xie, Cong;Mueller, Klaus

文献摘要

参考文献

被引文献

相似文献

深度学习的成功可以归功于人类专家数小时的参数和架构调整。神经架构搜索 (NAS) 技术旨在通过自动化 DNN 架构的搜索过程来解决这个问题,使非专家也能使用 DNN。具体来说,一次性 NAS 技术最近越来越受欢迎,因为众所周知,它们可以减少 NAS 技术的搜索时间。 One-Shot NAS 的工作原理是通过参数共享来训练一个大型模板网络,其中包括所有候选神经网络。接下来是通过评估随机选择的可能候选架构来应用程序对其组件进行排名。然而,随着这些搜索模型变得越来越强大和多样化,它们变得越来越难以理解。因此,即使搜索结果运行良好,也很难识别搜索偏差并控制搜索进程,因此需要可解释性和人机交互 (HIL) One-Shot NAS。为了缓解这些问题,我们推出了 NAS-Navigator,这是一种可视化分析 (VA) 系统,旨在解决 One-Shot NAS 的三个问题:与现有最先进 (SOTA) 技术相比,可解释性、HIL 设计和性能改进。 NAS-Navigator 将 NAS 的完全控制权交还给用户,同时仍然保留自动搜索的优势,从而为非专家用户提供帮助。分析师可以利用他们的领域知识并借助界面提示来指导搜索。评估结果证实我们改进的 One-Shot NAS 算法的性能与其他 SOTA 技术相当。使用 NAS-Navigator 添加可视化分析 (VA) 时,显示了搜索时间和性能的进一步改进。我们与几位深度学习研究人员合作设计了界面,并通过控制实验和专家访谈对 NAS-Navigator 进行了评估。
The success of DL can be attributed to hours of parameter and architecture tuning by human experts. Neural Architecture Search (NAS) techniques aim to solve this problem by automating the search procedure for DNN architectures making it possible for non-experts to work with DNNs. Specifically, One-shot NAS techniques have recently gained popularity as they are known to reduce the search time for NAS techniques. One-Shot NAS works by training a large template network through parameter sharing which includes all the candidate NNs. This is followed by applying a procedure to rank its components through evaluating the possible candidate architectures chosen randomly. However, as these search models become increasingly powerful and diverse, they become harder to understand. Consequently, even though the search results work well, it is hard to identify search biases and control the search progression, hence a need for explainability and human-in-the-loop (HIL) One-Shot NAS. To alleviate these problems, we present NAS-Navigator, a visual analytics (VA) system aiming to solve three problems with One-Shot NAS; explainability, HIL design, and performance improvements compared to existing state-of-the-art (SOTA) techniques. NAS-Navigator gives full control of NAS back in the hands of the users while still keeping the perks of automated search, thus assisting non-expert users. Analysts can use their domain knowledge aided by cues from the interface to guide the search. Evaluation results confirm the performance of our improved One-Shot NAS algorithm is comparable to other SOTA techniques. While adding Visual Analytics (VA) using NAS-Navigator shows further improvements in search time and performance. We designed our interface in collaboration with several deep learning researchers and evaluated NAS-Navigator through a control experiment and expert interviews.
BEAMES:回归任务的交互式多模型指导、选择和检查
DOI: 10.1109/mcg.2019.2922592
发表时间: 2019
影响因子: 1.8
作者:
Subhajit Das;Dylan Cashman;Remco Chang;A. Endert
通讯作者: A. Endert
DOI: --
发表时间: 2016-11
期刊: ArXiv
影响因子: --
作者:
Barret Zoph;Quoc V. Le
通讯作者: Barret Zoph;Quoc V. Le
DOI: 10.1136/ebmh.11.4.102
发表时间: 2008-10
期刊: Evidence Based Mental Health
影响因子: --
作者:
P. Cochat;L. Vaucoret;J. Sarles
通讯作者: P. Cochat;L. Vaucoret;J. Sarles
用于构建和验证回归模型的基于分区的框架
DOI: --
发表时间: 2013
影响因子: 5.2
作者:
T. Mühlbacher;H. Piringer
通讯作者: H. Piringer