Above-All
Welcome To My Blog! 👋👋👋
Version 2 (Last Updated Time: 2025/05/09).
For the Chinese version, please visit 中文版.
Introduction
This section shares the same content as the Intro page, providing an overview of the blog’s purpose, goals, and themes. Feel free to jump there for details.
Before The Blog
Three months ago, I stumbled upon Lilian Weng[1], the former VP of Research at OpenAI and a leading expert in GPT model safety. She formulated the famous Agent Recipe: Agent = LLM + Memory + Planning Skills + Tool Use.
In OpenAI’s exclusive interview with her, The Power of Continuous Learning[3], she said:
“I believe in the power of learning, and it is never too late to learn. Maintaining my personal blog[2] is a good way to keep this curiosity going and learn about new progress in the deep learning community regularly. I also encourage my team to keep on learning, whether related or unrelated to their current projects. Ideas in different topics or fields can often inspire new ideas and broaden the potential solution space.
I’m also a strong believer in teamwork. If everyone shines in their best strength, we will get 1+1 > 2. Meanwhile, we might often run into ‘dirty’ work, and personally, I’m very willing to take on those tasks because, as long as that’s the biggest blocker or that task can add the biggest value to the project, nothing should be considered ‘dirty’ or ‘trivial.’ I encourage people around me to do the same—being a team player and working together to expedite team productivity.”
This philosophy aligns perfectly with the Feynman Technique:
- Identify the subject. Write down everything you know about the topic.
- Teach it to a child. If you can explain a concept simply, you truly understand it.
- Identify knowledge gaps. This is where real learning happens.
- Organize, simplify, and tell a story. Structure your explanation clearly.
At the time, I had just entered university three months prior. Haunted by the remnants of high school’s exam-oriented mindset, I struggled to adapt to the vastness of knowledge in computer science. Without the rigid framework of standardized tests, how could I truly consolidate what I was learning?
Perhaps the answer was already before me.
I will create my own blog!
Whether through problem-solving, writing, or the Feynman Technique, the essence of learning lies in reinforcing knowledge through output. When you can fluently derive multiple proofs for a math problem, organize complex ideas into logical frameworks, or craft a blog post that enlightens others, you achieve effective intellectual output. This process not only solidifies understanding but also brings immense satisfaction.
Yet, writing—especially high-quality blog posts—is far more than mere typing.
Think back to writing argumentative essays in high school. Did you ever feel that frustrating disconnect between what you thought ≠ what you planned ≠ what you wrote ≠ what you truly meant to express? Often, what we call “inspiration” is just fragmented thoughts—far from a well-structured, insightful, and innovative article.
This is the real challenge of blogging—and its greatest value. How do we refine fleeting sparks of inspiration into polished prose? How do we balance logic and expertise to produce clear, compelling writing?
About Myself
For more details, see About.
I am Xiyuan Yang (杨希渊), an AI undergraduate at Shanghai Jiao Tong University, specializing in Reinforcement Learning, Multi-Agent Systems, and LLM Reasoning.
Contact:
About the Blog
Technical Setup
Xiyuan Yang’s Blog was launched on November 19, 2024.
This blog will evolve alongside my academic journey, sharing intriguing technical insights while honing my logical thinking and communication skills. Updates will be irregular in timing, content, and length.
Built with GitHub Pages and the Hexo static site generator.
Want your own blog? The tools are waiting!
Table of Contents
- Home
- Intro (You are here!)
- Archives (All posts in one place)
- Categories
- Tags
- About (More about me)
- Resume (Academic CV—under construction)
- Miscellaneous
- FAQ (Common questions)
- Friend Links (Peers’ blogs)
- All Posts (Update history)
- Daily Log (Learning notes)
- GitHub Intro
- Life Musings (Casual reflections)
Goals & Update Plans
See My Posts for the latest updates!
Current focus areas (Drafted in Dec 2024):
Update (May 2025): Surprisingly, most goals achieved!
- Programming Languages
- C/C++ ✅
- Python ✅ (Syntax, Anaconda, key libraries)
- HTML/JavaScript ❌
- Bash/Shell ✅
- Tools & Workflows
- LaTeX ✅
- CLI ✅
- Git ✅
- Vim ✅
- Linux & Servers (For deep learning)
- Linux commands ✅
- Tmux ✅
- SSH ✅
- [CS] Algorithms
- Data Structures ✅
- Algorithms ✅
- Core CS Fundamentals ❌
- AI Specialization
- Machine Learning ✅
- Reinforcement Learning ✅
- Deep Learning ✅ (PyTorch covered)
- AI Papers *(Classic & Cutting-edge) ✅
Abstract Math❌- Occasional Rants ✅
FAQ
Visit here for answers.
Friend Links
Discover blogs by peers and collaborators! Interested in connecting? Email me.
Resume
Future academic CV: Xiyuan Yang’s Resume.