Pilot
An introduction to Learning Rate
This is a pilot lesson that introduces essential concepts and provides hands-on tasks to reinforce learning, making it educational and interesting for beginners.
The Forward Pass 1
Neurons, Activation, and Layers
Explore the fundamental components of artificial neural networks! Topics covered include the structure and function of AI neurons, activation mechanisms, and the organization of neurons into layers to form neural networks.
The Forward Pass 2
Input/Hidden/Output Layers and Multi-Layer Perceptrons
Discover the core elements of Multi-Layer Perceptrons (MLPs). Delve into the architecture and operation of AI neurons, activation functions, and the layering process that constructs robust neural networks.
The Backward Pass 1
Loss Functions and Intro to Backpropagation
Delve into the concepts of loss functions, from theory to practice. Learn through examples and hands-on calculations. Then, unravel the Chain Rule's role in optimizing MLPs.
The Backward Pass 2
Computation of Backward Pass
Explore how computers calculate slopes to minimize loss. Visualize how these slopes move through layers step by step. Follow a simple example of the backward pass, demonstrating how slopes adjust model settings for improved learning.
The Bigger Picture
Network Training and Hyperparameters
Understand the process of dividing data into sets and the importance of Train/Dev/Test splits. Explore key concepts like learning rates, epochs, and learning rate decay, crucial for training models effectively. Visualize training and boundary on a Dot example, simplifying complex ideas into easy-to-understand visuals.