Deep Learning Slides
Explore the fundamentals, architectures, techniques, and real-world applications of Deep Learning.
Learn how Deep Learning enables machines to learn complex patterns from large datasets.
Understand neural networks, CNNs, RNNs, transformers, and essential training concepts.
Explore key techniques such as backpropagation, optimization, regularization, and transfer learning.
Discover practical applications of Deep Learning in computer vision, NLP, healthcare, and AI.
What You'll Learn
Introduction
Deep Learning vs Machine Learning — The Real Difference
A clear, visual guide to what actually separates deep learning from classical machine learning: han…
38 minThe Biological Neuron & the McCulloch-Pitts Model
Meet the first artificial neuron. This visual guide traces how the 1943 McCulloch-Pitts model turne…
32 minRosenblatt's Perceptron: The First Learning Machine
Go back to 1958, where deep learning began. This visual guide walks through Rosenblatt's perceptron…
37 minThe Multilayer Perceptron (MLP): Neural Networks from the Ground Up
Meet the workhorse of deep learning. This visual, animated guide walks through the MLP layer by lay…
37 minActivation Functions in Deep Learning: Sigmoid to GELU
Without activation functions, a hundred-layer network collapses into a single straight line. This v…
44 minForward Propagation in Neural Networks: A Visual Walkthrough
Watch an input flow through a network into a prediction. This visual guide breaks forward propagati…
31 minLoss Functions & Optimisation Objectives in ML
The loss function is a model's scoreboard — and training is the relentless push to shrink it. This …
38 minThe Backpropagation Algorithm: How Neural Networks Learn
Backpropagation is how a network turns one wrong guess into thousands of precise corrections. This …
33 minBackpropagation Solved Step by Step: A Worked Example
Follow one training step of a 2×2×1 network from start to finish — every weighted sum, activation, …
33 minBackpropagation Solved: 2×2×2 Numerical Example
A complete backpropagation walkthrough on a 2-input, 2-hidden, 2-output network. Every net value, a…
32 minRecurrent Neural Networks (RNN): Memory for Sequences
Language, music, and time series unfold in order — and RNNs are the networks built to read them one…
36 minDiscrete Convolution: The Math Behind CNNs
Slide a small grid of weights across an image, and at each stop record how strongly it matches — th…
35 minPooling & Spatial Hierarchy in CNNs
After convolution finds features, pooling summarises them — shrinking the map, keeping what matters…
35 minCNN Fully Solved Numericals: Conv → ReLU → Pool
Two complete CNN forward-pass problems worked entirely by hand. Watch a 5×5 and a 4×4 image flow th…
47 minPython Implementation of CNNs: Build, Train, Diagnose
Build a convolutional neural network from a blank script. This visual guide stacks conv blocks in K…
31 minLSTM Networks: From Cell State to Gates
Vanilla RNNs forget within a few steps. LSTMs add a protected cell-state "conveyor belt" and three …
34 minLong Short-Term Memory (LSTM) Networks
A complete tour of LSTM networks — the cell state and three gates, the four equations, why they bea…
35 minOptimizers in Deep Learning
A complete, visual tour of deep learning optimizers — from plain gradient descent through momentum,…
40 min