AI Knowledge Repository

Welcome to the AI section of my knowledge repo. The content here is structured to follow my personal learning path, which I think provided a solid foundation. Starting with the mathematical basics that I gathered from my maths courses during my studies at Imperial all the way through to a Deep Learning Specialization.

Maths, Statistics & Probability

AI is not magic - it’s fundamentally mathematics, so understanding the mathematical intricacies is important to grasping ML & AI's core. This section is dedicated to that.

Machine Learning

Machine learning is a branch of AI that can learn and adapt automatically with minimal human intervention, so it serves as our first real dive into AI application. The content in this subsection was initially based off courses provided by Frank Kane amongst others, and then built upon as I went on.

Deep Learning

Getting deeper into AI, we reach Deep Learning, a more specialized subset of machine learning. I started with the core elements of neural networks and their architectures - these fundamentals served as the basis for more complex models. As I progressed, I got into best practices, tricks of the trade, optimization techniques, etc.

Once I was familiar with the general principles, I explored specialized neural network architectures. These include Convolutional Neural Networks (CNNs) for tasks like image recognition, and Recurrent Neural Networks (RNNs) for sequence-based tasks like natural language processing or music synthesis.

The content in this section was initially based off courses by Andrew Ng on Coursera.

A Quick Overview

Even if AI, machine learning, and deep learning aren't your everyday topics of conversation, they've undoubtedly gained prominence in modern discussion. While the terms might sometimes seem synonymous, distinctions exist:

• Artificial Intelligence (AI): The broad realm of computer science dedicated to creating systems capable of performing tasks that would typically require human intelligence.
• Machine Learning (ML): A subset of AI, where systems can learn and improve from experience without being explicitly programmed.
• Deep Learning (DL): Situated within ML, it's a method that uses neural networks, often vast in scale, to simulate human-like learning.

As visualized in the diagram, imagine AI, ML, and DL as concentric circles. AI is the most broad, with ML inside it, and Deep Learning, the most specialized of the three, inside of ML. This is the order I have generally followed in my wiki.