A deep understanding of Deep Learning & Neural Networks with Python
A rigorous scientific and mathematical approach to mastering neural networks, moving from basic perceptrons to advanced architectures. Covers gradient descent, backpropagation, and specialized networks like CNNs and RNNs using an experimental learning style.
Instructor
Mike X Cohen
Duration
57.5 hours
The certificate
In progress — no credential yet
What it covered — 14 modules
- 01
Introduction
Welcome to the course! · Course philosophy and how to succeed · Download all course materials
4/4 lectures - 02
Python Intro: The Basics
Variables and Data Types · Lists, Tuples, and Dictionaries · Indexing and Slicing
6/6 lectures - 03
Foundations: Math, NumPy, and PyTorch
The Linear Algebra you need for Deep Learning · NumPy for fast numerical computing · Introduction to PyTorch Tensors
5/5 lectures - 04
Gradient Descent
The intuition of Gradient Descent · Math of Gradient Descent: Derivatives and Slopes · Stochastic, Batch, and Mini-batch Gradient Descent
5/5 lectures - 05
Artificial Neural Networks (ANNs)
The Perceptron: The building block · Multi-Layer Perceptrons (MLPs) · Activation Functions: Sigmoid, Tanh, ReLU, and Leaky ReLU
5/5 lectures - 06
Overfitting and Cross-Validation
The Bias-Variance Tradeoff · Training, Validation, and Test sets · K-fold Cross-Validation
0/4 lectures - 07
Regularization
L1 and L2 Regularization · Dropout: A simple way to prevent overfitting · Batch Normalization
0/4 lectures - 08
Feed-Forward Networks (FFNs) & Milestone Projects
FFN Architecture for Classification · FFN for Regression tasks · Milestone Project 1: Digit Recognition
0/4 lectures - 09
Autoencoders
Architecture of Autoencoders · Latent Space and Dimensionality Reduction · Denoising Autoencoders
0/4 lectures - 10
Convolutional Neural Networks (CNNs)
The Convolution operation: Kernels and Filters · Padding and Stride · Pooling layers: Max and Average pooling
0/5 lectures - 11
Transfer Learning
Why use Pre-trained models? · Fine-tuning vs. Fixed Feature Extractors · Using VGG, ResNet, and Inception in PyTorch
0/3 lectures - 12
Generative Adversarial Networks (GANs)
The Generator and Discriminator · The Min-Max Game · Training GANs: Challenges and Stability
0/4 lectures - 13
Recurrent Neural Networks (RNNs)
Modeling Sequential Data · Simple RNNs vs. LSTMs and GRUs · Natural Language Processing (NLP) basics with RNNs
0/3 lectures - 14
Ethics of Deep Learning
Bias in AI models · Deepfakes and Privacy · Sustainability and Compute costs
0/3 lectures
Toolkit from this course
More certificates