Welcome to HuangFuSL's blog¶
Table of contents¶
Blog timeline¶
变分自编码器(Variational Autoencoder,VAE)的结构和AE相似,都是使用编码器对输入进行压缩,再通过解码器重建。与AE不同的是,VAE是一个生成模型而不是编码模型,它反而和GAN中的生成器是一个原理。假设数据的生成是通过一个随机分布的隐变量\(z\)控制。即首先从一个容易采样的分布\(P_\theta(z)\)中采样出\(z\),再训练一个函数\(f(z)\),输出条件分布\(P_\theta(x | z)\),从其中采样得到\(x\)。据此,我们可以学习到输入数据\(x\)的概率分布\(P_\theta (x)\),根据输入样本最大化对数似然\(\log P_\theta(x)\)。
自编码器是一类无监督的学习方法,可以将输入数据压缩为低维的隐向量表示。它包含两部分网络结构,即编码器和解码器。编码器将输入数据压缩为隐向量,解码器尝试将隐向量重建为对应的数据。通过最小化解码器输出和输入数据之间的差异,编码器可以学习到输入数据的压缩表示。
Recent updates¶
- 2025-01-17:3095. 或值至少 K 的最短子数组 I,3097. 或值至少 K 的最短子数组 II
- 2025-01-15:3065. 超过阈值的最少操作数 I,3066. 超过阈值的最少操作数 II
- 2025-01-14:Variational AE
- 2025-01-13:2270. 分割数组的方案数,激活函数,注意力机制,BERT,Custom transformer,编码器与解码器,Llama-2,KV-Cache,损失函数,MLP,多臂老虎机,归一化,位置编码,transformer变种,transformer,置信上界算法
- 2025-01-12:2275. 按位与结果大于零的最长组
Currently working on¶
Customization¶
Click on the buttons to change the primary color.
Click on the buttons to change the accent color.
However, if you try to switch from dark mode to light mode or reversed, changes to the primary color and accent color will lose.
Building documentation¶
Run git clone https://github.com/HuangFuSL/HuangFuSL.github.io.git
to get the
source code.
Bootstrap icon installation¶
The site uses bootstrap icons, which are added as submodules in
third_party/icons
. You have to manually initialize the submodule.
git submodule update --recursive --remote
LaTeX support¶
The site uses xelatex
and dvisvgm
to render tex document to SVG images
embedded in the markdown files. However, as the SVG images are ignored by
.gitignore
, you have to manually perform the conversion.
For GitHub repository clones:
- Run
git submodule update --recursive --remote
to receive the template. - Make sure you have installed and correctly configured
xelatex
anddvisvgm
. - Add
./template
directory to$TEXINPUTS
environmental variable. - Execute
ci/convert.py
in the root directory of the repository. - Run
mkdocs serve
to view the images.
The template is located at HuangFuSL/latex-template
GitHub workflow¶
You need to install the dependencies stored in requirements.txt
before you can
start building the site:
pip install -r requirements.txt
There are cross-links in the site which require metadata defined in the page,
so the project should be built before mkdocs serve
is executed. The exported
metadata is saved in meta.json
after a build is successfully executed. To
build the site, execute the following command:
mkdocs build -d build
Execute mkdocs serve
, the built site will appear at http://127.0.0.1:8000
Acknowledgements¶
The blog relies on the following open-source projects:
The blog uses the following mkdocs plugins to function correctly.
- Neoteroi/mkdocs-plugins
- lukasgeiter/mkdocs-awesome-pages-plugin
- timvink/mkdocs-git-revision-date-localized-plugin
- zhaoterryy/mkdocs-git-revision-date-plugin
- squidfunk/mkdocs-material
- facelessuser/mkdocs-material-extensions
- fralau/mkdocs_macros_plugin
- danielfrg/mkdocs-jupyter
- prcr/mkdocs-meta-descriptions-plugin
Unless noted, content in this blog are shared under CC-BY-NC-SA 4.0 license.
Version information¶
commit 65f006ca2a56c1751c67df72b38fe3b529f26c37
Merge: 425cd429 b17bbe7a
Author: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Date: Fri Jan 17 15:21:36 2025 +0800
Merge pull request #446 from HuangFuSL/leetcode-250117
Update: Leetcode 250117