📖 Glossary

Every bit of jargon from the lessons, in plain English.

Activation function: A function applied to a neuron's weighted sum (e.g. ReLU, sigmoid) that adds non-linearity, letting stacked layers learn complex patterns.
Agent: An AI system that reasons and takes actions, often by calling tools in a loop, to complete a goal.
Attention: A mechanism that lets each token weigh and pull information from other tokens. The heart of the transformer.
Backpropagation: The algorithm that computes every weight's gradient by passing error signals backwards through a network using the chain rule.
Batch / mini-batch: A small group of training examples processed together in one optimisation step.
Bias (in data): Systematic skew in data or outcomes that can make a model unfair or unreliable for some groups.
Classification: Predicting a category, like spam vs. not spam.
CNN: Convolutional neural network. Uses sliding filters to detect local patterns, mainly in images.
Context window: How many tokens an LLM can consider at once, both prompt and response.
Diffusion model: A generative model that learns to turn noise into data by reversing a gradual noising process.
Embedding: A vector of numbers representing something (a word, image, user) so that similar things are close together.
Epoch: One full pass through the training dataset.
Fine-tuning: Further training a pre-trained model on a smaller, task-specific dataset.
Gradient: The slope of the loss with respect to each parameter; it points in the direction of steepest increase.
Gradient descent: An optimisation method that repeatedly steps parameters opposite to the gradient to reduce loss.
Hallucination: When a model confidently produces false or invented information.
Inference: Using a trained model to make predictions, with its parameters frozen.
Learning rate: The step size used in gradient descent.
LLM: Large language model. A big transformer trained on text to predict the next token.
LoRA: Low-Rank Adaptation. Fine-tunes by training small low-rank matrices while freezing the base weights.
Loss function: A number measuring how wrong a model's predictions are. Training minimises it.
Overfitting: When a model memorises training data, including its noise, and performs poorly on new data.
Parameters / weights: The learned numbers inside a model that determine its behaviour.
Precision: Of the items predicted positive, the fraction that truly are positive: TP / (TP + FP).
Prompt: The input text you give an LLM to steer its output.
Quantisation: Storing model weights with fewer bits (e.g. 4-bit) to save memory and speed up inference.
RAG: Retrieval-augmented generation. Fetches relevant documents and includes them in the prompt so answers are grounded.
Recall: Of the truly positive items, the fraction the model caught: TP / (TP + FN).
Regression: Predicting a continuous number, like a price or temperature.
Regularisation: Techniques (L2 penalty, dropout, etc.) that discourage overly complex models to reduce overfitting.
Reinforcement learning: Learning to act by trial and error to maximise reward from an environment.
RLHF: Reinforcement learning from human feedback. Uses human preference data to fine-tune model behaviour.
Temperature: A sampling setting controlling randomness. Lower is more deterministic, higher is more varied.
Token: A chunk of text (a word, part of a word or punctuation) that language models read and write.
Transformer: A neural network architecture built on attention, used by virtually all modern LLMs.
Validation set: Data held out from training to tune settings and check for overfitting.

Want more? Google's Machine Learning Glossary covers hundreds of terms.