PaperReading20200226
CanChen ggchen@mail.ustc.edu.cn
To share or not share
- Motivation: With the publiaction of NAS101, the author wants to study how weight sharing work in different search spaces and the correlation between the supernet and results returned by NAS101.
- Method: The paper did extensive experiments in weight sharing methods with different search spaces and showed that search space plays an important role in the correlation.
- Contribution: The work made good use of NAS101 but I did not see any novelty in this work.
NAS EVALUATION IS FRUSTRATINGLY HARD
- Motivation: NAS needs some criteria.
- Method: This paper did a wide range of experiments and showed that cell-based searched models tends to have similar accuracies.
Also, the selection of seeds and macro structures are important. The operations, instead, have less impact on the performance. - Contribution: This paper is a ILCR paper, probably because this paper analysed some important issues in NAS evaluation.
PaperReading20200226的更多相关文章
随机推荐
- cnpm - 解决 " cnpm : 无法加载文件 C:\Users\93457\AppData\Roaming\npm\cnpm.ps1,因为在此系统上禁止运行脚本。有关详细信息 。。。 "
1.在win10 系统中搜索框 输入 Windos PowerShell选择 管理员身份运行 2,打开了powershell命令行之后,输入 set-ExecutionPolicy RemoteSig ...
- RemoteView设置高度
刚开始内层LinearLayout直接用 android:layout_height="match_parent" <?xml version="1.0" ...
- 【Go语言系列】1.1、GO语言简介:什么是GO语言
一.Go的起源 Go语言的所有设计者都说,设计Go语言是因为 C++ 给他们带来了挫败感.在 Google I/O 2012 的 Go 设计小组见面会上,Rob Pike 是这样说的: 我们做了大量的 ...
- 操作系统OS - 阻塞(Blocking)非阻塞(Non-Blocking)与同步(Synchronous)异步(Asynchronous)
参考: http://blog.jobbole.com/103290/ https://www.zhihu.com/question/19732473/answer/23434554 http://b ...
- C# 集合类型学习
如果你掌握了一门语言的数据结构 ,那么你离掌握这门语言 也不远了 1.列表 对于list,值得一提的是 Capacity 属性,使用默认的构造函数 ,让我们用代码来说明 var intList = n ...
- Linux 目录结构与目录操作
目录结构 Linux的文件系统是采用级层式的树状目录结构,在此结构中的最上层是根目录"/",然后再次目录下再创建其他目录 在Linux系统中,一切皆文件 常见目录作用 / : 所有 ...
- PyQt5数据可视化
1.下载PyQtGraph模块 Windows上下载: pip install pyqtgraph Linux上下载: pip3 install pyqtgraph MacOS上下载: pip3 in ...
- 吴裕雄 PYTHON 神经网络——TENSORFLOW 学习率的设置
import tensorflow as tf TRAINING_STEPS = 10 LEARNING_RATE = 1 x = tf.Variable(tf.constant(5, dtype=t ...
- 吴裕雄 python 神经网络——TensorFlow 输入文件队列
import tensorflow as tf def _int64_feature(value): return tf.train.Feature(int64_list=tf.train.Int64 ...
- eclipse修改工作空间编码格式
一.修改workspace默认编码 eclipse打开window -> 打开preferences 二.修改jsp默认编码 eclipse打开window -> 打开preference ...