Alex Yu

Hello & Welcome

A Bit About Me

I'm a 16 year-old student from Toronto, Ontario. I've always been interested in finding solutions to different kinds of problems. Recently, I've been learning about artificial intelligence and neural nets and I'm super passionate about using deep learning to make a difference in the world!

I'm working on developing projects that leverage machine learning to tackle real world concepts. Right now, I'm exploring the applications of Convolutional Neural Networks and Recurrent Neural Networks in Computer Vision and Natural Language Processing and I hope you join in on my journey!

Machine Learning

Python and R for data science projects.

Digital Content

Adobe Photoshop and Premiere Pro for editing.

Game Development

Experience with Unity Engine for VR development.

Web Design

HTML, CSS, and JS for web pages and applications.


My Portfolio

  • All
  • Articles
  • Projects
  • Videos
  • PoseNet Camera Feed Demo
    An implementation in TensorFlow.js for real-time inference using data from the webcam to detect pose.
  • Generating Music With Artificial Intelligence
    What I learned from doing research on music generation with machine learning.
  • Getting A Machine To Do My English Homework For Me
    A high-level overview of Recurrent Neural Networks and Long Short Term Memory cells and how they can be used to write essays.
  • LSTM Nietzsche Text Generation
    Using RNNs and LSTMs that were trained on a corpus of writings by Nietzsche to sample and generate completely new sequences of text.
  • Using Neural Networks To Create Paintings
    An overview of style transfer and methods for real-time inference on mobile devices.
  • Why Failure Is Essential For Personal Growth
    Some of my key takeaways on adapting to failure from Ray Dalio’s Principles
  • Skin Lesion Classifier
    A model trained to classify seven different types of skin cancers.
  • Walmart Recommendations
    Suggestions to help Walmart Canada improve its customer experience while maintaining a low-cost operating model.
  • Titanic Survival Classifier
    A logistic regression model used to predict the likelihood of a person surviving the sinking of the Titanic.
  • How To Teach A Computer To See With Convolutional Neural Networks
    An article I wrote on how CNNs work, some real-world applications, and how to code one in Python.
  • MNIST Convolutional Neural Network
    The code for a CNN written in Keras that is trained on the MNIST dataset, a collection of 70,000 28x28 images of handwritten digits. The model achieves an accuracy of approximately 98 to 99 percent.

Contact Me

Feel free to reach out and ask me anything!