P-DARTS

2019-ICCV-Progressive Differentiable Architecture Search Bridging the Depth Gap Between Search and Evaluation

  • Tongji University && Huawei
  • GitHub: 200+ stars
  • Citation:49

Motivation

Question:

DARTS has to search the architecture in a shallow network while evaluate in a deeper one.

DARTS在浅层网络上搜索,在深层网络上评估(cifar search in 8-depth, eval in 20-depth)。

This brings an issue named the depth gap (see Figure 1(a)), which means that the search stage finds some operations that work well in a shallow architecture, but the evaluation stage actually prefers other operations that fit a deep architecture better.

Such gap hinders these approaches in their application to more complex visual recognition tasks.

Contribution

propose Progressive DARTS (P-DARTS), a novel and efficient algorithm to bridge the depth gap.

Bring two questions:

Q1: While a deeper architecture requires heavier computational overhead

we propose search space approximation which, as the depth increases, reduces the number of candidates (operations) according to their scores in the elapsed search process.

Q2:Another issue, lack of stability, emerges with searching over a deep architecture, in which the algorithm can be biased heavily towards skip-connect as it often leads to rapidest error decay during optimization, but, actually, a better option often resides in learnable operations such as convolution.

we propose search space regularization, which (i) introduces operation-level Dropout [25] to alleviate the dominance of skip-connect during training, and (ii) controls the appearance of skip-connect during evaluation.

Method

search space approximation

在初始阶段,搜索网络相对较浅,但是cell中每条边上的候选操作最多(所有操作)。在阶段 \(S_{k-1}\)中,根据学习到的网络结构参数(权值)来排序并筛选出权值(重要性)较高的 \(O_k\) 个操作,并由此搭建一个拥有 \(L_k\) 个cell的搜索网络用于下一阶段的搜索,其中, \(L_k > L_{k-1} , O_k < O_{k-1}\) .

这个过程可以渐进而持续地增加搜索网络的深度,直到足够接近测试网络深度。

search space regularization

we observe that information prefers to flow through skip-connect instead of convolution or pooling, which is arguably due to the reason that skip-connect often leads to rapid gradient descent.

实验结果表明在本文采用的框架下,信息往往倾向于通过skip-connect流动,而不是卷积。这是因为skip-connect通常处在梯度下降最速的路径上。

the search process tends to generate architectures with many skip-connect operations, which limits the number of learnable parameters and thus produces unsatisfying performance at the evaluation stage.

在这种情况下,最终搜索得到的结构往往包含大量的skip-connect操作,可训练参数较少,从而使得性能下降。

We address this problem by search space regularization, which consists of two parts.

First, we insert operation-level Dropout [25] after each skip-connect operation, so as to partially ‘cut off’ the straightforward path through skip-connect, and facilitate the algorithm to explore other operations.

作者采用搜索空间正则来解决这个问题。一方面,作者在skip-connect操作后添加Operations层面的随机Dropout来部分切断skip-connect的连接,迫使算法去探索其他的操作。

However, if we constantly block the path through skip-connect, the algorithm will drop them by assigning low weights to them, which is harmful to the final performance.

然而,持续地阻断这些路径的话会导致在最终生成结构的时候skip-connect操作仍然受到抑制,可能会影响最终性能。

we gradually decay the Dropout rate during the training process in each search stage, thus the straightforward path through skip-connect is blocked at the beginning and treated equally afterward when parameters of other operations are well learned, leaving the algorithm itself to make the decision.

因此,作者在训练的过程中逐渐地衰减Dropout的概率,在训练初期施加较强的Dropout,在训练后期将其衰减到很轻微的程度,使其不影响最终的网络结构参数的学习。

Despite the use of Dropout, we still observe that skip-connect, as a special kind of operation, has a significant impact on recognition accuracy at the evaluation stage.

另一方面,尽管使用了Operations层面的Dropout,作者依然观察到了skip-connect操作对实验性能的强烈影响。

This motivates us to design the second regularization rule, architecture refinement, which simply controls the number of preserved skip-connects, after the final search stage, to be a constant M.

因此,作者提出第二个搜索空间正则方法,即在最终生成的网络结构中,保留固定数量的skip-connect操作。具体的,作者根据最终阶段的结构参数,只保留权值最大的M个skip-connect操作,这一正则方法保证了搜索过程的稳定性。在本文中, M=2 .

We emphasize that the second regularization technique must be applied on top of the first one, otherwise, in the situations without operation-level Dropout, the search process is producing low-qualityarchitectureweights, basedon which we could not build up a powerful architecture even with a fixed number of skip-connects.

需要强调的是,第二种搜索空间正则是建立在第一种搜索空间正则的基础上的。在没有执行第一种正则的情况下,即使执行第二种正则,算法依旧会生成低质量的网络结构。

Experiments

Cell arch in different Search Stage

cifar10

ImageNet

·

Conclusion

we propose a progressive version of differentiable architecture search to bridge the depth gap between search and evaluation scenarios.

The core idea is to gradually increase the depth of candidate architectures during the search process.

  • 2Q: computational overhead and instability

Search space approximate and Search space regularize

Our research defends the importance of depth in differentiable architecture search, depth is still the dominant factor in exploring the architecture space.

2019-ICCV-PDARTS-Progressive Differentiable Architecture Search Bridging the Depth Gap Between Search and Evaluation-论文阅读的更多相关文章

  1. 论文笔记:Progressive Differentiable Architecture Search:Bridging the Depth Gap between Search and Evaluation

    Progressive Differentiable Architecture Search:Bridging the Depth Gap between Search and Evaluation ...

  2. 论文笔记:DARTS: Differentiable Architecture Search

    DARTS: Differentiable Architecture Search 2019-03-19 10:04:26accepted by ICLR 2019 Paper:https://arx ...

  3. 论文笔记:Progressive Neural Architecture Search

    Progressive Neural Architecture Search 2019-03-18 20:28:13 Paper:http://openaccess.thecvf.com/conten ...

  4. (转)Illustrated: Efficient Neural Architecture Search ---Guide on macro and micro search strategies in ENAS

    Illustrated: Efficient Neural Architecture Search --- Guide on macro and micro search strategies in  ...

  5. 2019-ICLR-DARTS: Differentiable Architecture Search-论文阅读

    DARTS 2019-ICLR-DARTS Differentiable Architecture Search Hanxiao Liu.Karen Simonyan.Yiming Yang GitH ...

  6. 2019 ICCV、CVPR、ICLR之视频预测读书笔记

    2019 ICCV.CVPR.ICLR之视频预测读书笔记 作者 | 文永亮 学校 | 哈尔滨工业大学(深圳) 研究方向 | 视频预测.时空序列预测 ICCV 2019 CVP github地址:htt ...

  7. 微软的一篇ctr预估的论文:Web-Scale Bayesian Click-Through Rate Prediction for Sponsored Search Advertising in Microsoft’s Bing Search Engine。

    周末看了一下这篇论文,觉得挺难的,后来想想是ICML的论文,也就明白为什么了. 先简单记录下来,以后会继续添加内容. 主要参考了论文Web-Scale Bayesian Click-Through R ...

  8. LeetCode 33 Search in Rotated Sorted Array [binary search] <c++>

    LeetCode 33 Search in Rotated Sorted Array [binary search] <c++> 给出排序好的一维无重复元素的数组,随机取一个位置断开,把前 ...

  9. 33. Search in Rotated Sorted Array & 81. Search in Rotated Sorted Array II

    33. Search in Rotated Sorted Array Suppose an array sorted in ascending order is rotated at some piv ...

随机推荐

  1. IDEA中如何使用debug调试项目 一步一步详细教程

    转载该文章:https://blog.csdn.net/yxl_1207/article/details/80973622 一.Debug开篇 首先看下IDEA中Debug模式下的界面. 如下是在ID ...

  2. 关于Cookie的一点简单认识

    1.Cookie Cookies是服务器在本地机器上存储的小段文本并随每一个请求发送至同一服务器,是在客户端保持状态的方案.通常每个 Cookie 的大小不能超过4KB.客户端每次向服务器发出请求,就 ...

  3. 【Scala】利用Akka的actor编程模型,实现2个进程间的通信

    文章目录 步骤 一.创建maven工程,导入jar包 二.master进程代码开发 三.worker进程代码开发 四.控制台结果 步骤 一.创建maven工程,导入jar包 <propertie ...

  4. IO 模型知多少 | 代码篇

    引言 之前的一篇介绍IO 模型的文章IO 模型知多少 | 理论篇 比较偏理论,很多同学反应不是很好理解.这一篇咱们换一个角度,从代码角度来分析一下. socket 编程基础 开始之前,我们先来梳理一下 ...

  5. linux下安装gmp遇到 configure:error:no usable m4 in$path or /user/5bin解决方案

    安装过程中遇到如下报错: 上面的报错是因为你没有安装m4,安装m4就可以了:以下两种命令人选其一: #yum install m4 或 #apt-get install m4 ps:如果遇到权限问题就 ...

  6. css 箭头三角形

    1.向下的三角形 .down{ display:inline-block; width:0px; height:0px; border-top:8px solid rgba(0, 0, 0, 0.65 ...

  7. [Unity3D]编辑器扩展之数组或List显示

    效果如下: 源码如下: using System.Collections.Generic; using UnityEditor; using UnityEngine; namespace XM.Edi ...

  8. nginx操作目录

    nginx配置文件/conf/nginx.conf 错误日志功能:los/error.log 访问日志功能:logs/access.log 站点服务请求功能配置:html/ 禁止访问功能配置 404页 ...

  9. jquery VS Dom(小实例单选-多选-反选)

    一直以来大家对jquery评价莫过于六个字 “吃得少,干的多” ,应用实例让大家看看这款牛到爆的插件能帮我们做什么,话不多说,直接加码 <!DOCTYPE html> <html l ...

  10. Redis学习笔记(十三) 复制(下)

    上一篇写了Redis复制功能的简单应用,下面我们看下Redis复制功能的实现过程.下面基本上是理论部分,枯燥乏味,但希望大家能看看,毕竟知识不都是感兴趣的.耐得住寂寞,经得起诱惑,方能守得住繁华 ~. ...