<![CDATA[SINGULARITY]]>

<![CDATA[SINGULARITY]]>https://blog.youkuaiyun.com/UNOboroszh-cnhttps://blog.youkuaiyun.com/<![CDATA[GRPO本质上就是REINFORCE的特例（对上了！这下都对上了）]]>https://blog.youkuaiyun.com/UNOboros/article/details/145400083https://blog.youkuaiyun.com/UNOboros/article/details/145400083UNOborosThu, 30 Jan 2025 19:52:21 +0800<![CDATA[一种局部优化TRPO的思路]]>https://blog.youkuaiyun.com/UNOboros/article/details/140660547https://blog.youkuaiyun.com/UNOboros/article/details/140660547UNOborosWed, 24 Jul 2024 15:48:40 +0800<![CDATA[从RL的专业角度解惑 instruct GPT的目标函数]]>https://blog.youkuaiyun.com/UNOboros/article/details/140264512https://blog.youkuaiyun.com/UNOboros/article/details/140264512UNOborosMon, 08 Jul 2024 13:18:40 +0800<![CDATA[理解策略梯度方法：从REINFORCE到PPO]]>https://blog.youkuaiyun.com/UNOboros/article/details/140187335https://blog.youkuaiyun.com/UNOboros/article/details/140187335UNOborosThu, 04 Jul 2024 20:06:01 +0800<![CDATA[注意力机制的一般框架]]>https://blog.youkuaiyun.com/UNOboros/article/details/136189893https://blog.youkuaiyun.com/UNOboros/article/details/136189893UNOborosTue, 20 Feb 2024 15:45:09 +0800<![CDATA[根式为什么是根式？（我可能疯了？）]]>https://blog.youkuaiyun.com/UNOboros/article/details/126243493https://blog.youkuaiyun.com/UNOboros/article/details/126243493UNOborosTue, 09 Aug 2022 11:53:07 +0800<![CDATA[对换子的深层含义]]>https://blog.youkuaiyun.com/UNOboros/article/details/125974489https://blog.youkuaiyun.com/UNOboros/article/details/125974489UNOborosMon, 25 Jul 2022 14:51:32 +0800<![CDATA[一种直观理解Galois理论的途径]]>https://blog.youkuaiyun.com/UNOboros/article/details/125958044https://blog.youkuaiyun.com/UNOboros/article/details/125958044UNOborosSun, 24 Jul 2022 16:11:10 +0800<![CDATA[关于围棋围住的最大面积问题]]>https://blog.youkuaiyun.com/UNOboros/article/details/125245159https://blog.youkuaiyun.com/UNOboros/article/details/125245159UNOborosSun, 12 Jun 2022 14:39:51 +0800<![CDATA[关于协变矢量和逆变矢量的直观理解]]>https://blog.youkuaiyun.com/UNOboros/article/details/79765228https://blog.youkuaiyun.com/UNOboros/article/details/79765228UNOborosSat, 31 Mar 2018 05:13:55 +0800<![CDATA[混入Assasin的海盗分金问题]]>https://blog.youkuaiyun.com/UNOboros/article/details/72905076https://blog.youkuaiyun.com/UNOboros/article/details/72905076UNOborosWed, 07 Jun 2017 23:33:18 +0800<![CDATA[关于Mobius环的一个小思想实验]]>https://blog.youkuaiyun.com/UNOboros/article/details/65442412https://blog.youkuaiyun.com/UNOboros/article/details/65442412UNOborosThu, 23 Mar 2017 13:31:45 +0800<![CDATA[火车上的醉汉问题（一个非常深奥的问题）]]>https://blog.youkuaiyun.com/UNOboros/article/details/53431427https://blog.youkuaiyun.com/UNOboros/article/details/53431427UNOborosFri, 02 Dec 2016 09:23:31 +0800<![CDATA[计算的极限——关于Diophantine以及黎曼猜想]]>https://blog.youkuaiyun.com/UNOboros/article/details/51199125https://blog.youkuaiyun.com/UNOboros/article/details/51199125UNOborosWed, 20 Apr 2016 11:15:19 +0800<![CDATA[变化之理]]>https://blog.youkuaiyun.com/UNOboros/article/details/50560726https://blog.youkuaiyun.com/UNOboros/article/details/50560726UNOborosFri, 22 Jan 2016 13:10:54 +0800<![CDATA[Lob定理的证明（觉得自己集合论学得好其实是一种幻觉……）]]>https://blog.youkuaiyun.com/UNOboros/article/details/50379333https://blog.youkuaiyun.com/UNOboros/article/details/50379333UNOborosTue, 22 Dec 2015 13:36:30 +0800<![CDATA[一种可能的新的人脑信息网络模型]]>https://blog.youkuaiyun.com/UNOboros/article/details/50365065https://blog.youkuaiyun.com/UNOboros/article/details/50365065UNOborosSun, 20 Dec 2015 19:01:40 +0800<![CDATA[关于Hadoop以及S3的几个坑]]>https://blog.youkuaiyun.com/UNOboros/article/details/50173669https://blog.youkuaiyun.com/UNOboros/article/details/50173669UNOborosFri, 04 Dec 2015 13:41:22 +0800<![CDATA[Description of graph6 and sparse6 encodings]]>https://blog.youkuaiyun.com/UNOboros/article/details/49757905https://blog.youkuaiyun.com/UNOboros/article/details/49757905UNOborosTue, 10 Nov 2015 12:19:52 +0800<![CDATA[最终之LMS]]>https://blog.youkuaiyun.com/UNOboros/article/details/49384539https://blog.youkuaiyun.com/UNOboros/article/details/49384539UNOborosSat, 24 Oct 2015 13:59:36 +0800