Skip to content
View hien078's full-sized avatar
2 followers · 2 following

Block or report hien078

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

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

Report abuse
hien078/README.md

Hi, I'm Hien 👋

🎓 Applied Mathematics student at USTH (University of Science and Technology of Hanoi) 🔬 AI / ML Researcher & Mathematical Software Engineer — bridging rigorous mathematics and modern machine learning 📐 Building knowledge from first principles: derive mathematically, code from scratch, and verify empirically.

🌟 Flagship Project

🤖

Machine-Learning-from-scratch

22 ML & Deep Learning algorithms — derived by hand, implemented in pure NumPy & PyTorch, verified by tests.

CI Docs License: MIT Python

  • 22 topics in five phases: OLS → SVM → Random Forest → CNN → LSTM → Transformer → VAE/Diffusion → RL → GNN → LLM engineering (BPE/LoRA/DPO) → SSL
  • Every topic = theory.md derivation + from-scratch notebook + exercises, backed by an installable library with 65 unit tests, 85% coverage floor, and CI-executed notebooks
  • 📖 Read it as a website: hien078.github.io/Machine-Learning-from-scratch

The rest of the first-principles trilogy:

  • 🧮 applied-mathematics-foundation — the math prerequisites: Linear Algebra, Calculus, Probability & Statistics, Optimization, Information Theory, ODEs, Graph Theory, Numerical Methods
  • 📐 first-principles-math-modeling — 15 modeling topics: dynamical systems, SIR epidemics, game theory, LP/IP/KKT optimization, simulation — 121 Colab-ready notebooks

🔬 Research & Engineering Focus

  • Mathematical Foundations of Deep Learning: Convex & non-convex optimization, matrix calculus, spectral theory.
  • Neural ODEs & Dynamical Systems: Physics-Informed Neural Networks (PINNs), continuous-time models.
  • First-Principles ML: Building transparent, scalable ML algorithms from scratch without black-box abstractions.
  • Information & Optimization Theory: Variational inference, ELBO bounds, information geometry.

🛠️ Tech Stack & Tools

  • Languages & Frameworks: Python PyTorch NumPy SciPy Pandas scikit-learn
  • Math & Authoring: LaTeX KaTeX JupyterLab Matplotlib Plotly
  • DevOps & OS: Git GitHub Actions GitHub CLI Linux (Bash) VS Code

📊 GitHub Stats

Hien's GitHub Stats Top Languages

"The purpose of computing is insight, not numbers." — Richard Hamming

Popular repositories Loading

  1. first-principles-math-modeling first-principles-math-modeling Public

    📐 First-principles mathematical modeling — 15 topics: dynamical systems, SIR epidemic models, ODEs, optimization, game theory, simulation & data-driven modeling. All with Jupyter notebooks.

    Jupyter Notebook 1

  2. applied-mathematics-foundation applied-mathematics-foundation Public

    🧮 Rigorous applied mathematics for AI & Data Science — 8 comprehensive modules: Linear Algebra, Calculus, Probability & Statistics, Optimization, Information Theory, ODEs, Graph Theory, Numerical M…

    Jupyter Notebook 1

  3. hien078 hien078 Public

  4. postiz-app postiz-app Public

    Forked from gitroomhq/postiz-app

    📨 The ultimate agentic social media scheduling tool 🤖

    TypeScript

  5. VideoLingo VideoLingo Public

    Forked from Huanshere/VideoLingo

    Netflix-level subtitle cutting, translation, alignment, and even dubbing - one-click fully automated AI video subtitle team | Netflix级字幕切割、翻译、对齐、甚至加上配音，一键全自动视频搬运AI字幕组

    Python

  6. MoneyPrinterTurbo MoneyPrinterTurbo Public

    Forked from harry0703/MoneyPrinterTurbo

    利用 AI 大模型和自动化工作流，根据主题或关键词一键生成高清短视频。Generate HD short videos from a topic or keyword with an automated AI workflow.

    Python