Skip to content
View guoxun's full-sized avatar

Block or report guoxun

Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
guoxun/README.md

Hi there, I'm Guo Xun. πŸ‘‹

I work on search and recommendation, LLM algorithms, machine learning, and recommender systems. My long-term focus is on building practical and scalable intelligent systems that connect retrieval, ranking, causal inference, and product-facing algorithm design.

🀝 Research Collaboration

I am open to discussions and collaboration on topics related to:

  • Search, ads, and recommendation algorithms
  • Recommender systems and ranking optimization
  • Causal inference for industrial ML systems
  • Data mining and practical machine learning
  • LLM applications for search and recommendation

If you are exploring similar directions, feel free to reach out.


🌱 Research Interests

My interests center on making algorithm systems more useful in real production environments. I am especially interested in the following directions:

πŸ”Ž 1. Search and Recommendation Systems

Search and recommendation systems are where algorithm research directly meets product value. I care about:

  • Retrieval, ranking, and matching
  • Recommendation strategy optimization
  • User modeling and personalization
  • Large-scale recommendation system evolution

βš–οΈ 2. Causal Inference and Data Mining

To make decision systems more reliable, I pay close attention to:

  • Causal inference in recommendation and ranking
  • Evaluation and decision-making under bias
  • High-value signal mining from large-scale data

πŸ€– 3. LLM and Applied Machine Learning

As foundation models become part of industrial stacks, I am interested in:

  • LLM techniques for search and recommendation
  • Bridging classical ML and modern foundation models
  • Turning algorithm ideas into deployable systems

πŸ’‘ Mission

My goal is to bridge algorithm research, industrial recommendation systems, and knowledge sharing, so that machine learning methods can better serve the continuous evolution of search and recommendation products.

I care not only about model quality, but also about whether a method is explainable, deployable, and useful in practice.


🧭 Background & Experience

  • πŸŽ“ Academic Training
    I studied Computer Science and Technology at Beijing Jiaotong University from 2011 to 2015, and later pursued joint training in Computer Science and Technology at Peking University from 2015 to 2017.

  • πŸ’Ό Industry Experience
    I have worked in algorithm-related roles across Huawei, Alibaba Group, and Baidu, with a continuing focus on search, recommendation, and machine learning systems.


πŸ’‘ Writing & Code

πŸ“š Knowledge Sharing
On Zhihu, I write about topics including recommendation algorithms, recommender systems, search and recommendation, causal inference, and computational biology.

As of March 9, 2026, my public Zhihu profile shows:

  • 10,549 followers
  • 173 answers
  • 26 articles

I also run the WeChat public account: ζœΊε™¨ε­¦δΉ δΈŽε•†δΈšζ™Ίθƒ½ε‰ζ²Ώ.

🧠 Selected GitHub Projects

  • AI_game: browser game experiments built with Codex
  • Editor-Qt: a game editor similar to RPG Maker
  • pytorch-vdsr: PyTorch implementation of VDSR
  • Lipreading_DBN: visual speech recognition using Deep Belief Network
  • PRMLT: MATLAB implementations of machine learning algorithms from PRML

πŸ“« Get in Touch

Popular repositories Loading

  1. pytorch-vdsr pytorch-vdsr Public

    Forked from twtygqyy/pytorch-vdsr

    VDSR (CVPR2016) pytorch implementation

    Jupyter Notebook 1

  2. hello-world hello-world Public

    one step of me,one step of the world haha

  3. HearthStone-Calculator HearthStone-Calculator Public

    a calculator of HearthStone,it can help you calculate your damage to get victory.

  4. learngit learngit Public

  5. gitskills gitskills Public

  6. 2015_study_II 2015_study_II Public