Modern AI in Ten Topics
From algorithms to large language models — the sequence of ideas that explains where AI is in the mid-2020s and how it actually works.
- Reading time
- ~31 min (+23 min optional)
- Level mix
- 2 beginner · 14 intermediate · 1 advanced
Modern AI didn’t appear from nowhere. It’s the latest layer on top of decades of work on algorithms, statistics, neural networks, and accelerator hardware. This path takes you from “what’s an algorithm” all the way to “what’s a large language model”, with enough machine learning in between to make the last few topics make sense.
By the end you should be able to read a paper abstract, follow a technical blog post about LLMs, and know what questions to push back on when someone claims their AI product does something extraordinary.
Roadmap
Loading progress...
Background
A precise, finite recipe for solving a problem — the central idea of computer science.
- Data StructureOptional
A way of organising values in memory so that the operations you care about — find, insert, delete, sort — are efficient.
Machine learning
Building systems that improve at a task by learning from data instead of being explicitly programmed for every case.
Learning a function from labelled examples — the most widely-deployed flavour of machine learning.
The two distinct phases of a machine-learning model's life — learning its parameters (training) and using them to make predictions (inference).
The optimisation algorithm that trains almost every neural network — iteratively nudge each parameter in the direction that reduces the loss.
The algorithm that computes how much each weight in a neural network contributed to the error — applying the chain rule layer by layer in reverse so gradient descent knows which way to adjust every parameter.
Deep learning
A family of machine learning models loosely inspired by the brain — layers of simple units that, together, can approximate complex functions.
- Convolutional Neural NetworkOptional
A neural-network architecture that uses convolution to detect local patterns — edges, textures, shapes — in grid-like data such as images; the workhorse of computer vision for a decade.
The neural-network architecture, introduced in 2017, that powers modern large language models, image generation, and most of contemporary AI.
A neural-network technique that lets a model weigh which parts of its input matter most for each part of its output — the key idea behind transformers and modern language models.
A very large neural network — usually a transformer — trained on huge amounts of text to predict the next token. The basis of modern chat assistants and AI coding tools.
- EmbeddingOptional
A learned vector representation of an item — a word, an image, a user, a product — where geometric distance roughly equals semantic similarity.
Specialised learning
- Reinforcement LearningOptional
A branch of machine learning where an agent learns to act in an environment by trial-and-error, optimising a reward signal.
- Computer VisionOptional
The field of teaching computers to interpret images and video — classification, detection, segmentation, generation.
- Natural Language ProcessingOptional
The field of getting computers to understand, generate, and work with human language — from spam filters and translation to the large language models behind modern chatbots.
Hardware
- GPUOptional
A processor designed for massive data-parallel work — originally for rendering graphics, now also the workhorse of machine learning, simulation, and crypto.