Deep Learning (DL) · Beginner level

Beginner level · Module 1 of 5 · 10 guided hours

Neural Network Foundations

Understand artificial neurons, layers and activations, how a network makes a prediction and how it learns by gradient descent.

Learning objectives

  • Calculate a neuron's output from inputs, weights and bias
  • Describe the forward pass and loss of a network
  • Explain gradient descent and backpropagation intuitively

Topics

  • neurons, weights and bias
  • activation functions
  • layers and the forward pass
  • loss functions
  • gradient descent
  • backpropagation intuition

Lessons

  1. Lesson 1 Neurons, Weights and Activations

    See how an artificial neuron combines inputs with weights and a bias and passes the result through an activation function.

    weights and biasweighted sumReLUsigmoid

  2. Lesson 2 Layers and the Forward Pass

    Connect neurons into layers and follow data through a network to a prediction and a loss.

    input, hidden and output layersfully connected layerforward passloss

  3. Lesson 3 How Networks Learn: Gradient Descent

    Understand how networks adjust their weights to reduce the loss, step by step.

    gradientgradient descentbackpropagationepoch and batch

Practical activity

Compute a tiny network's output by hand, then build the same network in PyTorch and confirm the result.

Expected outcomes

  • A clear mental model of how networks compute and learn
  • First working PyTorch code

Assessment

Hand-calculation quiz and a short notebook reproducing it in PyTorch.

Free taster lesson Introduction to Artificial Neural Networks (ANNs) Read a free self-study lesson on this topic in the Free Course Library.

Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.