Madry A, Makelov A, Schmidt L, et al. Towards Deep Learning Models Resistant to Adversarial Attacks.[J]. arXiv: Machine Learning, 2017.

@article{madry2017towards,

title={Towards Deep Learning Models Resistant to Adversarial Attacks.},

author={Madry, Aleksander and Makelov, Aleksandar and Schmidt, Ludwig and Tsipras, Dimitris and Vladu, Adrian},

journal={arXiv: Machine Learning},

year={2017}}

利用特定的方法产生"坏"样本(Adversarial samples), 以此来促进网络的稳定性是当下的热点之一, 本文以实验为主, 比较PGD( projected gradient descent) 和 FGSM(fast gradient sign method)在不同数据下的表现, 以及由普通样本产生"坏"样本会出现的一些现象.

主要内容

Adversarial attacks 主要聚焦于下列问题:

\[\tag{2.1}
\min_{\theta} \rho (\theta) \quad where \quad \rho(\theta) =\mathbb{E}_{(x,y)\sim D}[\max_{\delta \in S} L(\theta, x+\delta, y)].
\]

其中\(S\)是我们指定的摄动集合, 直接一点就是\(|\delta| <constant\)之类.

通过FGSM产生"坏"样本:

\[x + \epsilon \: \mathrm{sgn}(\nabla_x L(\theta,x,y)).
\]

这个思想是很直接的(从线性感知器谈起, 具体看here).

PGD的思路是, 给定摄动集\(S\), 比如小于某个常数的摄动(e.g. \(\{\tilde{x}:\|x-\tilde{x}\|_{\infty}<c\}\)), 多次迭代寻找合适的adversarial samples:

\[x^{t+1} = \prod_{x+S} (x^t + \alpha \: \mathrm{sgn} (\nabla_x L(\theta,x, y)),
\]

其中\(\prod\)表示投影算子, 假设\(S=\{\tilde{x}:\|x-\tilde{x}\|_{\infty}<c\}\),

\[x^{t+1} = \arg \min_{z \in x+S} \frac{1}{2} \|z - (x^t + \alpha \: \mathrm{sgn} (\nabla_x L(\theta,x, y))\|_2^2,
\]

实际上, 可以分开讨论第\((i,j)\)个元素, \(y:=(x^t + \alpha \: \mathrm{sgn} (\nabla_x L(\theta,x, y))\), 只需找到\(z_{ij}\)使得

\[\|z_{ij}-y_{ij}\|_2
\]

最小即可. 此时有显示解为:

\[z_{ij}=
\left \{
\begin{array}{ll}
x_{ij} +c & y_{ij} > x_{ij}+c \\
x_{ij} -c & y_{ij} < x_{ij}-c \\
y_{ij} & else.
\end{array} \right.
\]

简而言之就是一个截断.

重复几次, 至到\(x^t\)被判断的类别与初始的\(x\)不同或者达到最大迭代次数.

Note

  • 如果我们训练网络能够免疫PGD的攻击, 那么其也能很大一部分其它的攻击.
  • FGSM对抗训练不能提高网络的稳定性(在摄动较大的时候).
  • weak models may fail to learn non-trival classfiers.
  • 网络越强(参数等程度)训练出来的稳定性越好, 同时可转移(指adversarial samples 在多个网络中被误判)会变差.

Towards Deep Learning Models Resistant to Adversarial Attacks的更多相关文章

  1. How to Grid Search Hyperparameters for Deep Learning Models in Python With Keras

    Hyperparameter optimization is a big part of deep learning. The reason is that neural networks are n ...

  2. a Javascript library for training Deep Learning models

    w强化算法和数学,来迎接机器学习.神经网络. http://cs.stanford.edu/people/karpathy/convnetjs/ ConvNetJS is a Javascript l ...

  3. Run Your Tensorflow Deep Learning Models on Google AI

    People commonly tend to put much effort on hyperparameter tuning and training while using Tensoflow& ...

  4. What are some good books/papers for learning deep learning?

    What's the most effective way to get started with deep learning?       29 Answers     Yoshua Bengio, ...

  5. (转) Awesome Deep Learning

    Awesome Deep Learning  Table of Contents Free Online Books Courses Videos and Lectures Papers Tutori ...

  6. (转)分布式深度学习系统构建 简介 Distributed Deep Learning

    HOME ABOUT CONTACT SUBSCRIBE VIA RSS   DEEP LEARNING FOR ENTERPRISE Distributed Deep Learning, Part ...

  7. The Brain vs Deep Learning Part I: Computational Complexity — Or Why the Singularity Is Nowhere Near

    The Brain vs Deep Learning Part I: Computational Complexity — Or Why the Singularity Is Nowhere Near ...

  8. Paper Reading——LEMNA:Explaining Deep Learning based Security Applications

    Motivation: The lack of transparency of the deep  learning models creates key barriers to establishi ...

  9. Coursera Deep Learning 2 Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - week1, Assignment(Regularization)

    声明:所有内容来自coursera,作为个人学习笔记记录在这里. Regularization Welcome to the second assignment of this week. Deep ...

随机推荐

  1. Scala和Java的List集合互相转换

    import java.util import scala.collection.mutable /** * 集合互相转换 */ object ScalaToJava { def main(args: ...

  2. 强化学习实战 | 表格型Q-Learning玩井字棋(二)

    在 强化学习实战 | 表格型Q-Learning玩井字棋(一)中,我们构建了以Game() 和 Agent() 类为基础的框架,本篇我们要让agent不断对弈,维护Q表格,提升棋力.那么我们先来盘算一 ...

  3. Ecshop 后台导出订单Excel时, 内存溢出的解决方法

    今天继续跟大家分享一下,在我配置Ecshop时的问题. 今天的问题是在后台想要导出订单列表Excel时出现的内存溢出.错误提示如下 问题:  Fatal error: Allowed memory s ...

  4. 3.1 go context代码示例

    context.WithCancel返回两个有关联的对象,ctx与cancel,调用cancel发送一个空struct给ctx,ctx一旦接收到该对象后,就终止goroutine的执行;ctx是线程安 ...

  5. String类型和包装类型作为参数传递时,是属于值传递还是引用传递呢?

    原理知识: 如果参数类型是原始类型,那么传过来的就是这个参数的一个副本,也就是这个原始参数的值,这个跟之前所谈的传值是一样的.如果在函数中改变了副本的 值不会改变原始的值. 如果参数类型是引用类型,那 ...

  6. MySQL批量数据脚本示例

    一.建表 # 新建库 create database bigData; use bigData; #1 建表dept CREATE TABLE dept( id INT UNSIGNED PRIMAR ...

  7. Delphi编译报错对照表

    ';' not allowed before 'ELSE' → ElSE前不允许有";" " clause not allowed in OLE automation s ...

  8. IT服务生命周期

    一.概述 IT服务生命周期由规划设计(Pianning&Design).部署实施(Implementing).服务运营(Opera,tion).持续改进(Improvemenit)和监督管理( ...

  9. [BUUCTF]PWN——pwnable_orw

    pwnable_orw 附件 步骤: 例行检查,32位程序,开启了canary 本地运行一下程序,看看大概的情况,提示我们输入shellcode 32位ida载入,检索字符串,没看见什么可以直接利用的 ...

  10. libevent 源码分析

    1,前言 Libevent是一个轻量级的开源高性能网络库,使用者众多,研究者更甚,相关文章也不少.写这一系列文章的用意在于,一则分享心得:二则对libevent代码和设计思想做系统的.更深层次的分析, ...