Back to Academy

Neural Foundations

The bedrock of modern intelligence. Before you build, you must understand the mathematical elegance of the artificial neuron.

01

Linear Algebra & Tensors

Everything in a neural network is a matrix. Learn how vectors represent data and how matrix multiplication fuels the forward pass.

Key Concepts

  • • Vector Spaces
  • • Dot Products
  • • Matrix Transformations

Resources

  • • 3Blue1Brown Linear Algebra
  • • MIT 18.06 Course
02

Calculus & Optimization

How does a network learn? Through gradients. Understand partial derivatives, the chain rule, and Gradient Descent.

Key Concepts

  • • Loss Functions
  • • Backpropagation
  • • Learning Rates

Resources

  • • Khan Academy Calculus
  • • DeepLearning.ai Math Specialization
    • 03

    The Perceptron

    The atom of AI. Learn about weights, biases, and activation functions like ReLU, Sigmoid, and Tanh.

    Key Concepts

    • • Weighted Sums
    • • Non-linearity
    • • Decision Boundaries

    Resources

  • • Neural Networks and Deep Learning (Nielsen)

    • Ready to implement?

    Once you've mastered the theory, it's time to write your first tensor operation.

    Next: Start Coding