<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[egglikeFe的博客]]></title><description><![CDATA[]]></description><link>https://blog.csdn.net/egglikeFe</link><language>zh-cn</language><generator>https://blog.csdn.net/</generator><copyright><![CDATA[Copyright &copy; egglikeFe]]></copyright><item><title><![CDATA[浅聊RLVR]]></title><link>https://blog.csdn.net/egglikeFe/article/details/150770850</link><guid>https://blog.csdn.net/egglikeFe/article/details/150770850</guid><author>egglikeFe</author><pubDate>Mon, 25 Aug 2025 19:41:50 +0800</pubDate><description><![CDATA[首先，RL的根本是提高采样效率，但是传统的RL需要依赖人类标注或者复杂的奖励模型，但是RLVR改变了这点，他不需要上述的条件，而是通过规则化验证器生成二元反馈（0/1）。根据任务的类型，可验证类型分为三类：1、正确性验证：通过模型输出和标准答案做匹配来生成奖励。2、执行验证：通过代码解释器执行生成的代码，根据单元测试结果提供奖励。3、可验证约束：强制模型遵循输出格式或者拒绝不当的请求，违法规则奖励则为0.]]></description><category></category></item><item><title><![CDATA[从MHA、GQA、MQA到MHLA，注意力的几种处理方式]]></title><link>https://blog.csdn.net/egglikeFe/article/details/145584328</link><guid>https://blog.csdn.net/egglikeFe/article/details/145584328</guid><author>egglikeFe</author><pubDate>Wed, 12 Feb 2025 09:12:56 +0800</pubDate><description><![CDATA[主要介绍几种注意力处理方式，包括多头注意力、分组查询注意力、多查询注意力、多头潜在注意力~]]></description><category></category></item><item><title><![CDATA[基于BiLSTM_CRF的实体抽取]]></title><link>https://blog.csdn.net/egglikeFe/article/details/145028119</link><guid>https://blog.csdn.net/egglikeFe/article/details/145028119</guid><author>egglikeFe</author><pubDate>Thu, 09 Jan 2025 11:04:58 +0800</pubDate><description><![CDATA[简述基于BiLSTM_CRF的实体抽取方法~]]></description><category></category></item><item><title><![CDATA[NEO4J安装]]></title><link>https://blog.csdn.net/egglikeFe/article/details/144751829</link><guid>https://blog.csdn.net/egglikeFe/article/details/144751829</guid><author>egglikeFe</author><pubDate>Thu, 26 Dec 2024 20:02:09 +0800</pubDate><description><![CDATA[主要讲述neo4j这个工具的安装方法~]]></description><category></category></item><item><title><![CDATA[迁移学习--fasttext概述]]></title><link>https://blog.csdn.net/egglikeFe/article/details/144592564</link><guid>https://blog.csdn.net/egglikeFe/article/details/144592564</guid><author>egglikeFe</author><pubDate>Thu, 19 Dec 2024 19:12:58 +0800</pubDate><description><![CDATA[主要介绍了fasttext的三种使用方式~]]></description><category></category></item><item><title><![CDATA[Transformer简述和实现]]></title><link>https://blog.csdn.net/egglikeFe/article/details/144300281</link><guid>https://blog.csdn.net/egglikeFe/article/details/144300281</guid><author>egglikeFe</author><pubDate>Fri, 06 Dec 2024 20:58:28 +0800</pubDate><description><![CDATA[主要是讲解transformer的架构和实现~]]></description><category></category></item><item><title><![CDATA[循环神经网络（RNN）简述]]></title><link>https://blog.csdn.net/egglikeFe/article/details/144138136</link><guid>https://blog.csdn.net/egglikeFe/article/details/144138136</guid><author>egglikeFe</author><pubDate>Fri, 29 Nov 2024 15:22:09 +0800</pubDate><description><![CDATA[RNN及其变体，包括LSTM和GRU，以及注意力的计算方式~]]></description><category></category></item><item><title><![CDATA[文本的预处理(pytorch)]]></title><link>https://blog.csdn.net/egglikeFe/article/details/144039257</link><guid>https://blog.csdn.net/egglikeFe/article/details/144039257</guid><author>egglikeFe</author><pubDate>Mon, 25 Nov 2024 20:29:41 +0800</pubDate><description><![CDATA[本文主要讲解文本预处理的几种手段，包括分词，文本的张量表示，文本数据分析，文本特征分析，文本特征处理和文本特征增强。]]></description><category></category></item><item><title><![CDATA[安装Pytorch和cuda指导]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143905510</link><guid>https://blog.csdn.net/egglikeFe/article/details/143905510</guid><author>egglikeFe</author><pubDate>Wed, 20 Nov 2024 11:05:46 +0800</pubDate><description><![CDATA[介绍pytorch与cuda的安装~]]></description><category></category></item><item><title><![CDATA[卷积神经网络基础]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143864532</link><guid>https://blog.csdn.net/egglikeFe/article/details/143864532</guid><author>egglikeFe</author><pubDate>Mon, 18 Nov 2024 19:20:04 +0800</pubDate><description><![CDATA[本文主要介绍了卷积神经网络的基础知识~]]></description><category></category></item><item><title><![CDATA[深度学习--神经网络基础]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143723064</link><guid>https://blog.csdn.net/egglikeFe/article/details/143723064</guid><author>egglikeFe</author><pubDate>Tue, 12 Nov 2024 20:27:46 +0800</pubDate><description><![CDATA[介绍包括激活函数、初始化方法、神经网络搭建、损失函数、网络优化方法以及正则化方法。]]></description><category></category></item><item><title><![CDATA[pytorch基础语法]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143628814</link><guid>https://blog.csdn.net/egglikeFe/article/details/143628814</guid><author>egglikeFe</author><pubDate>Fri, 08 Nov 2024 16:02:03 +0800</pubDate><description><![CDATA[主要介绍pytorch一些基础语法~]]></description><category></category></item><item><title><![CDATA[支持向量机SVM简述]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143286184</link><guid>https://blog.csdn.net/egglikeFe/article/details/143286184</guid><author>egglikeFe</author><pubDate>Mon, 28 Oct 2024 10:40:56 +0800</pubDate><description><![CDATA[简述机器学习中的支持向量机SVM~]]></description><category></category></item><item><title><![CDATA[Kmeans聚类算法简述]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143209039</link><guid>https://blog.csdn.net/egglikeFe/article/details/143209039</guid><author>egglikeFe</author><pubDate>Thu, 24 Oct 2024 14:46:49 +0800</pubDate><description><![CDATA[Kmeans相关概述，包括四种系数SSE，SC，SH系数以及肘方法~]]></description><category></category></item><item><title><![CDATA[简述特征降维的几种方式]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143090286</link><guid>https://blog.csdn.net/egglikeFe/article/details/143090286</guid><author>egglikeFe</author><pubDate>Sun, 20 Oct 2024 11:35:45 +0800</pubDate><description><![CDATA[主要讲解低方差过滤法，主成分分析PCA和相关系数法~]]></description><category></category></item><item><title><![CDATA[机器学习中的朴素贝叶斯]]></title><link>https://blog.csdn.net/egglikeFe/article/details/143029242</link><guid>https://blog.csdn.net/egglikeFe/article/details/143029242</guid><author>egglikeFe</author><pubDate>Thu, 17 Oct 2024 21:46:12 +0800</pubDate><description><![CDATA[对机械学习中的朴素贝叶斯做简单叙述]]></description><category></category></item><item><title><![CDATA[决策树和集成学习的概念以及部分推导]]></title><link>https://blog.csdn.net/egglikeFe/article/details/142902922</link><guid>https://blog.csdn.net/egglikeFe/article/details/142902922</guid><author>egglikeFe</author><pubDate>Sun, 13 Oct 2024 19:56:35 +0800</pubDate><description><![CDATA[决策树：ID3，C4.5，CART决策树和剪枝
集成学习：随机森林，adaboost，GBDT，XGBoost]]></description><category></category></item><item><title><![CDATA[机器学习入门]]></title><link>https://blog.csdn.net/egglikeFe/article/details/142797147</link><guid>https://blog.csdn.net/egglikeFe/article/details/142797147</guid><author>egglikeFe</author><pubDate>Wed, 09 Oct 2024 21:35:11 +0800</pubDate><description><![CDATA[机械学习入门，讲解一些基础知识~]]></description><category></category></item><item><title><![CDATA[回溯算法--python]]></title><link>https://blog.csdn.net/egglikeFe/article/details/142695750</link><guid>https://blog.csdn.net/egglikeFe/article/details/142695750</guid><author>egglikeFe</author><pubDate>Thu, 03 Oct 2024 20:32:00 +0800</pubDate><description><![CDATA[本文主要讲述了回溯算法，讨论了几种常见的场景全排列问题、子集和问题和N皇后问题~]]></description><category></category></item><item><title><![CDATA[分治思想--python]]></title><link>https://blog.csdn.net/egglikeFe/article/details/142621911</link><guid>https://blog.csdn.net/egglikeFe/article/details/142621911</guid><author>egglikeFe</author><pubDate>Sat, 28 Sep 2024 21:19:30 +0800</pubDate><description><![CDATA[本文主要讲解了分治的解题思路，以二分查找和恢复二叉树为基础来理解分治~]]></description><category></category></item></channel></rss>