<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[MXZJSJ1112的博客]]></title><description><![CDATA[我相信，我终将成功！！！]]></description><link>https://blog.youkuaiyun.com/MXZJSJ1112</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; MXZJSJ1112]]></copyright><item><title><![CDATA[强化学习-随机近似与随机梯度下降]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145816524</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145816524</guid><author>MXZJSJ1112</author><pubDate>Tue, 04 Mar 2025 23:54:00 +0800</pubDate><description><![CDATA[本篇博客主要包含如下内容：1️⃣ 新的角度再看mean eatimation；3️⃣ 随机梯度下降以及它的各种变体（GD、BGD、SGD、MBGD）。]]></description><category></category></item><item><title><![CDATA[文科生都能学会的-coze搭建《论文辅助阅读器》-V1]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145938916</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145938916</guid><author>MXZJSJ1112</author><pubDate>Sat, 01 Mar 2025 00:53:47 +0800</pubDate><description><![CDATA[文科生都能操作的coze搭建AI应用开发！！！]]></description><category></category></item><item><title><![CDATA[【论文分享02】A brief introduction to weakly supervised learning]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145862870</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145862870</guid><author>MXZJSJ1112</author><pubDate>Tue, 25 Feb 2025 21:33:25 +0800</pubDate><description><![CDATA[本论文主要介绍了弱监督学习，探讨其在数据标注成本高场景下的应用，概述了相关研究的进展，为后续研究提供了参考。下面是论文的脑图。监督学习在有强监督信息时成果显著，但实际任务中获取监督信息成本高，弱监督学习愈发重要。文章主要聚焦不完全、不确切和不精确监督，而实际中多种类型常同时出现且还有其他类型的弱监督，感兴趣的同学们可以搜索相关的资料进行阅读。]]></description><category></category></item><item><title><![CDATA[迎接2025，立个flag]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145802143</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145802143</guid><author>MXZJSJ1112</author><pubDate>Sat, 22 Feb 2025 22:45:44 +0800</pubDate><description><![CDATA[2025计划书]]></description><category></category></item><item><title><![CDATA[【论文分享01】Communication is All You Need: Persuasion Dataset Construction via Multi-LLM Communication]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145714990</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145714990</guid><author>MXZJSJ1112</author><pubDate>Tue, 18 Feb 2025 21:36:23 +0800</pubDate><description><![CDATA[该论文提出了一种多LLM通信框架，用于自动生成高质量、多样化的说服性对话数据，为说服相关研究提供了新途径和有价值的资源。下面是论文的脑图。]]></description><category></category></item><item><title><![CDATA[DeepSeek-R1技术报告快速解读]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145482208</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145482208</guid><author>MXZJSJ1112</author><pubDate>Sun, 09 Feb 2025 21:19:40 +0800</pubDate><description><![CDATA[本篇是对DeepSeek-R1技术论文的简单整理，辅助大家快速了解DeepSeek-R1的相关技术背景。]]></description><category></category></item><item><title><![CDATA[【kimi k1.5技术报告】KIMI K1.5: SCALING REINFORCEMENT LEARNING WITH LLMS]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145340666</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/145340666</guid><author>MXZJSJ1112</author><pubDate>Fri, 24 Jan 2025 14:54:26 +0800</pubDate><description><![CDATA[本文对Kimi K1.5的技术报告进行了简单的解读，供大家快速阅读和思考！！！]]></description><category></category></item><item><title><![CDATA[强化学习-蒙特卡洛方法]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144905661</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144905661</guid><author>MXZJSJ1112</author><pubDate>Thu, 16 Jan 2025 23:45:57 +0800</pubDate><description><![CDATA[本篇博客开始介绍model-free的方法，model-free的核心思想是基于数据来估计出一个模型，核心包括：蒙特卡洛思想、MC Basic算法、MC Exploring Starts算法以及MC Epsilon-Greedly算法。]]></description><category></category></item><item><title><![CDATA[【论文阅读-思维链的构造方法02】4.1.2 Automatic Construction-02]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144695601</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144695601</guid><author>MXZJSJ1112</author><pubDate>Thu, 09 Jan 2025 14:14:29 +0800</pubDate><description><![CDATA[本篇涉及4篇论文，是上一篇4.1.2 Automatic Construction-01的延续，旨在解决如何自动构建CoT。]]></description><category></category></item><item><title><![CDATA[【代码随想录-数组篇02】：双指针(快慢指针)法相关力扣练习题]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144965328</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144965328</guid><author>MXZJSJ1112</author><pubDate>Wed, 08 Jan 2025 16:42:08 +0800</pubDate><description><![CDATA[本篇从力扣5道编程题出发，为大家展示双指针的用法。]]></description><category></category></item><item><title><![CDATA[【代码随想录-数组篇01】：二分查找相关力扣练习题]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144896450</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144896450</guid><author>MXZJSJ1112</author><pubDate>Fri, 03 Jan 2025 16:44:18 +0800</pubDate><description><![CDATA[本篇共包含5道使用二分查找可以解决的力扣编程题目，全部用python语言进行实践。]]></description><category></category></item><item><title><![CDATA[【论文阅读-思维链的构造方法02】4.1.2 Automatic Construction-01]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144643270</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144643270</guid><author>MXZJSJ1112</author><pubDate>Thu, 02 Jan 2025 16:32:57 +0800</pubDate><description><![CDATA[本文是思维链综述中 4.1.2 Automatic Construction小结中涉及的前4篇论文，从三方面进行讲解：1）提出了什么问题？2）如何解决的？3）效果及展望。]]></description><category></category></item><item><title><![CDATA[强化学习-值迭代与策略迭代(动态规划)]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144609155</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144609155</guid><author>MXZJSJ1112</author><pubDate>Mon, 23 Dec 2024 22:09:36 +0800</pubDate><description><![CDATA[本文针对强化学习第4课的内容进行了总结，包含3部分：value iteration；policy iteration；truncated policy iteration。]]></description><category></category></item><item><title><![CDATA[Docker基础知识]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144422153</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144422153</guid><author>MXZJSJ1112</author><pubDate>Sun, 22 Dec 2024 21:04:29 +0800</pubDate><description><![CDATA[docker基础知识]]></description><category></category></item><item><title><![CDATA[【论文阅读-思维链的构造方法01】4.1.1 Manual Prompting]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144229720</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144229720</guid><author>MXZJSJ1112</author><pubDate>Thu, 19 Dec 2024 11:08:46 +0800</pubDate><description><![CDATA[【思维链综述】第四章节中-4.1.1 Manual Prompting小节涉及的5篇论文要点总结与方法对比。]]></description><category></category></item><item><title><![CDATA[强化学习-贝尔曼最优公式]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144481949</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144481949</guid><author>MXZJSJ1112</author><pubDate>Mon, 16 Dec 2024 00:32:29 +0800</pubDate><description><![CDATA[贝尔曼最优公式数学理论与求解]]></description><category></category></item><item><title><![CDATA[Shell脚本基础语法教程]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144421214</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144421214</guid><author>MXZJSJ1112</author><pubDate>Thu, 12 Dec 2024 11:12:52 +0800</pubDate><description><![CDATA[本文简单记录了Shell脚本的相关知识。预计阅读时间10分钟+。]]></description><category></category></item><item><title><![CDATA[强化学习-贝尔曼公式]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144007970</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/144007970</guid><author>MXZJSJ1112</author><pubDate>Tue, 10 Dec 2024 22:24:24 +0800</pubDate><description><![CDATA[本文主要包含以下部分：1）return的重要性；2）state value和action value的定义以及之间的关系；3）贝尔曼公式的推导。]]></description><category></category></item><item><title><![CDATA[【思维链综述】-Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances,]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/143850110</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/143850110</guid><author>MXZJSJ1112</author><pubDate>Fri, 22 Nov 2024 14:48:56 +0800</pubDate><description><![CDATA[大模型技术  ----> 思维链]]></description><category></category></item><item><title><![CDATA[强化学习-基本概念]]></title><link>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/143839144</link><guid>https://blog.youkuaiyun.com/MXZJSJ1112/article/details/143839144</guid><author>MXZJSJ1112</author><pubDate>Mon, 18 Nov 2024 23:20:55 +0800</pubDate><description><![CDATA[强化学习基本概念]]></description><category></category></item></channel></rss>