<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[Hacker_ccc的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/Hacker_ccc</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; Hacker_ccc]]></copyright><item><title><![CDATA[CTF-密码学基础]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/138591868</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/138591868</guid><author>Hacker_ccc</author><pubDate>Thu, 09 May 2024 01:52:05 +0800</pubDate><description><![CDATA[本文主要介绍密码学的一些基础，介绍了古典密码体系，现代密码(主要是MD5算法和RSA算法)，最后结合CTF中的题目进行了一些介绍]]></description><category></category></item><item><title><![CDATA[CTF-隐写术]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/138414841</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/138414841</guid><author>Hacker_ccc</author><pubDate>Fri, 03 May 2024 13:53:58 +0800</pubDate><description><![CDATA[CTF隐写术(Steganography)]]></description><category></category></item><item><title><![CDATA[卷积神经网络CNN]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127739127</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127739127</guid><author>Hacker_ccc</author><pubDate>Mon, 07 Nov 2022 21:11:30 +0800</pubDate><description><![CDATA[本文详细介绍了卷积神经网络CNN的原理和步骤，从为什么需要CNN的思考入手，深入理解CNN存在的意义；并详细讲解了每个步骤，卷积步骤的具体实现，池化步骤的具体实现，如何利用keras框架实现CNN等。从神经网络的可解释性方面引入了Deep Dream和Deep Style，当下时兴有趣的东西；最后将CNN推广至更多的应用，如当下热门的Alpha Go，语音识别，文本挖掘等等方面。还有一些扩展文章供读者学习。]]></description><category></category></item><item><title><![CDATA[训练模型的技巧方法]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127513632</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127513632</guid><author>Hacker_ccc</author><pubDate>Tue, 25 Oct 2022 15:11:04 +0800</pubDate><description><![CDATA[本文介绍了训练神经网络时，训练集表现不好或者测试集表现不好时，如何根据不同的大体情况，调节模型，使得模型表现得更好。里面介绍了新的激活函数，重点介绍了ReLU函数，涉及到了先前梯度下降法就提到过的自适应学习率；同时介绍了Early Stopping，正则化和DropOut等方法来调节过拟合的情况。]]></description><category></category></item><item><title><![CDATA[Keras框架认识]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127363748</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127363748</guid><author>Hacker_ccc</author><pubDate>Mon, 17 Oct 2022 14:58:38 +0800</pubDate><description><![CDATA[本文中简单引入了keras，同时对mini batch的设置进行了解释和分析，最后对经典的手写数字辨识的问题，进行了代码示例。]]></description><category></category></item><item><title><![CDATA[BP反向传播网络]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127306544</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127306544</guid><author>Hacker_ccc</author><pubDate>Thu, 13 Oct 2022 18:32:18 +0800</pubDate><description><![CDATA[本文介绍了如何通过反向传播误差修正模型参数，从梯度下降法等原理处学习如何进行反向传播，进而了解为什么模型参数的修正和激活函数相关。]]></description><category></category></item><item><title><![CDATA[分类判别式模型——逻辑斯特回归曲线]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127197580</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127197580</guid><author>Hacker_ccc</author><pubDate>Fri, 07 Oct 2022 19:00:49 +0800</pubDate><description><![CDATA[本文介绍了分类的判别式模型，从以往机器学习的三大步骤引入；在寻找最优解中，比较了与线性回归梯度下降法的不同；在损失函数层面，比较了交叉熵和square error的差异；在分类模型上，比较是本专栏上文的分类生成模型。最后在多分类问题上进行了扩展，在无法解决的同或问题中引入了特征映射和神经网络的概念。]]></description><category></category></item><item><title><![CDATA[分类——生成模型]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127078228</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/127078228</guid><author>Hacker_ccc</author><pubDate>Tue, 27 Sep 2022 20:50:24 +0800</pubDate><description><![CDATA[通过概率生成模型，充分理解朴素贝叶斯分类过程中的整体过程，其中不缺乏对数学的理解，并通过数学模型，推导出逻辑斯特回归曲线模型。该文章可以使你对分类的概念更上一层楼]]></description><category></category></item><item><title><![CDATA[梯度下降法]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126939854</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126939854</guid><author>Hacker_ccc</author><pubDate>Mon, 19 Sep 2022 19:21:37 +0800</pubDate><description><![CDATA[本文介绍了机器学习中常用于解决优化问题的梯度下降法；不局限于传统的梯度下降，文章还介绍了几种自适应式的梯度下降法，每个参数基于不同的学习率去进行梯度下降；同时讲到了Stochastic的梯度下降和数据的规范化，一定程度上能够提高效率；最后通过数学层面的分析，充分理解为什么梯度下降能够起作用。文章最后还提及了梯度下降法的一些限制。]]></description><category></category></item><item><title><![CDATA[偏差与方差]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126918342</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126918342</guid><author>Hacker_ccc</author><pubDate>Sun, 18 Sep 2022 15:40:36 +0800</pubDate><description><![CDATA[从概率论、统计学的原始角度，充分理解偏差与方差；对不同的函数model的bias和Variance有充分理解，便于后续工作（模型选择，超参数调优）的一个开展。]]></description><category></category></item><item><title><![CDATA[ML class Note——回归]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126909881</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126909881</guid><author>Hacker_ccc</author><pubDate>Sat, 17 Sep 2022 19:53:11 +0800</pubDate><description><![CDATA[课程笔记，用于复习]]></description><category></category></item><item><title><![CDATA[PageRank算法]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126593860</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126593860</guid><author>Hacker_ccc</author><pubDate>Mon, 29 Aug 2022 22:22:18 +0800</pubDate><description><![CDATA[介绍了随机游走算法中常用的PageRank算法的基本过程，枚举了一些例子便于更好的理解；同时针对PageRank的问题，提出引入一个α进行修正。]]></description><category></category></item><item><title><![CDATA[Degree and Closeness Centrality]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126332252</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126332252</guid><author>Hacker_ccc</author><pubDate>Sun, 14 Aug 2022 15:39:30 +0800</pubDate><description><![CDATA[从入度和接近性定义了网络中节点的中心性，并引入了规约对多个连通分量的问题进行了修正解决]]></description><category></category></item><item><title><![CDATA[Network robustness]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126332211</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126332211</guid><author>Hacker_ccc</author><pubDate>Sun, 14 Aug 2022 15:34:00 +0800</pubDate><description><![CDATA[介绍了网络的稳健性/鲁棒性，找到割边或割点]]></description><category></category></item><item><title><![CDATA[Connected component]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126332173</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126332173</guid><author>Hacker_ccc</author><pubDate>Sun, 14 Aug 2022 15:30:39 +0800</pubDate><description><![CDATA[介绍图的连通分量，针对无向图和有向图，图连通分量的不同计算方法]]></description><category></category></item><item><title><![CDATA[Distance Measures]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126294380</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126294380</guid><author>Hacker_ccc</author><pubDate>Thu, 11 Aug 2022 22:20:03 +0800</pubDate><description><![CDATA[定义了网络中的距离和路径，通过BFS解决无权最短路问题，并引进了直径、偏心率、半径等新概念，设定了周边和中心节点的概念；并引入了经典网络Karate Club Network进行示例]]></description><category></category></item><item><title><![CDATA[Clustering Coefficient]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126273673</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126273673</guid><author>Hacker_ccc</author><pubDate>Wed, 10 Aug 2022 20:19:44 +0800</pubDate><description><![CDATA[关于网络中聚类系数的计算——LCC和GCC的计算方法，GCC的计算方法涉及平均LCC和Transitivity两种，并比较了一下差异]]></description><category></category></item><item><title><![CDATA[机器学习之集成学习]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126240765</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126240765</guid><author>Hacker_ccc</author><pubDate>Tue, 09 Aug 2022 09:23:25 +0800</pubDate><description><![CDATA[本文将为你介绍集成学习，从最原始的Bagging到随机森林，最后是Boosting分类，涉及到权重分配式的AdaBoost和竞赛万金油算法XGBoost]]></description><category></category></item><item><title><![CDATA[神经网络分类]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126190667</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126190667</guid><author>Hacker_ccc</author><pubDate>Sat, 06 Aug 2022 10:32:09 +0800</pubDate><description><![CDATA[从感知机出发，讲到多层感知机到多层神经网络，同时涉及常用的误差反向传播网络，即常说的BP算法，普及了其公式，并引入简单明了的例子更好的理解这个算法。最后是对sklearn多层感知机api的学习...]]></description><category></category></item><item><title><![CDATA[聚类算法实践部分]]></title><link>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126141986</link><guid>https://blog.youkuaiyun.com/Hacker_ccc/article/details/126141986</guid><author>Hacker_ccc</author><pubDate>Wed, 03 Aug 2022 15:50:40 +0800</pubDate><description><![CDATA[本文介绍的是聚类算法的实践，文章设计了层次聚类和聚类质量的评估方法；例如利用霍普金斯统计量估计聚类趋势，利用肘方法SSE确定聚类和轮廓系数检验聚类质量。最后有通过API实现Kmeans和层次聚类的处理...]]></description><category></category></item></channel></rss>