A complete AI curriculum, in the right order
Follow the tracks from top to bottom for a full degree-style journey, or jump straight to the area you need. Each track builds on the mathematics and ideas of the ones before it.
01
AI Foundations
Search, logic, knowledge, uncertainty and the ideas that started the field.
- 01What Is Artificial Intelligence? A Rigorous Introduction
- 02A History of AI: From Turing to Transformers
- 03The Turing Test and Its Critics
- 04Intelligent Agents, Environments and the PEAS Framework
- 05Uninformed Search: BFS, DFS, Uniform-Cost and Iterative Deepening
- 06Informed Search: Greedy Best-First and A* with Admissible Heuristics
- 07Adversarial Search: Minimax and Alpha–Beta Pruning
- 08Constraint Satisfaction Problems: Backtracking, Propagation and Heuristics
- 09Propositional Logic for AI: Syntax, Semantics and Inference
- 10First-Order Logic: Objects, Relations, Quantifiers and Unification
- 11Knowledge Representation: Semantic Networks, Frames, Ontologies and Knowledge Graphs
- 12Expert Systems and Rule-Based AI: Rise, Fall and Legacy
- 13Bayesian Networks: Reasoning Under Uncertainty
- 14Hidden Markov Models: Filtering, Smoothing and the Viterbi Algorithm
- 15Classical Planning: STRIPS, PDDL and Planning Graphs
- 16Symbolic vs Connectionist AI — and the Neuro-Symbolic Synthesis
- 17Local Search: Hill Climbing, Simulated Annealing and Beam Search
- 18Genetic Algorithms and Evolutionary Computation
- 19Fuzzy Logic: Reasoning with Degrees of Truth
- 20Monte Carlo Tree Search: The Algorithm Behind Superhuman Go
- 21Decision Theory: Utility, Expected Value and the Value of Information
- 22Swarm Intelligence: Particle Swarm Optimisation and Ant Colony Optimisation
- 23Narrow AI, General AI and Superintelligence: Separating Science from Speculation
- 24The Landscape of Modern AI: A Map for Students
02
Mathematics for ML
Linear algebra, calculus, probability, statistics, information theory and optimisation.
- 01Why Mathematics Matters for Machine Learning
- 02Vectors and Vector Spaces: The Language of Data
- 03Matrices and Linear Transformations
- 04Eigenvalues and Eigenvectors: The Natural Axes of a Transformation
- 05Singular Value Decomposition: The Swiss Army Knife of Linear Algebra
- 06Norms, Inner Products, Projections and Distances
- 07Derivatives, Gradients and the Chain Rule
- 08Matrix Calculus Essentials: Jacobians, Hessians and Vector Derivatives
- 09Probability Fundamentals: Sample Spaces, Axioms and Conditional Probability
- 10Random Variables and Probability Distributions
- 11Bayes' Theorem: Updating Beliefs with Evidence
- 12Expectation, Variance, Covariance and Correlation
- 13The Gaussian Distribution: Why It Is Everywhere
- 14Maximum Likelihood Estimation: How Models Learn from Data
- 15MAP Estimation, Priors and the Bayesian View of Regularisation
- 16Information Theory for ML: Entropy, Cross-Entropy and KL Divergence
- 17Convex Optimisation Basics: Why Some Problems Are Easy
- 18Gradient Descent: Theory, Step Sizes and Convergence
- 19Lagrange Multipliers and Constrained Optimisation (KKT Conditions)
- 20Statistical Hypothesis Testing for ML: Is Model A Really Better?
- 21Sampling and Monte Carlo Methods
- 22Markov Chains: Memoryless Processes and Stationary Distributions
- 23Numerical Stability: Floating Point, Log-Sum-Exp and Avoiding NaNs
- 24The Curse of Dimensionality
- 25Tensors and Tensor Operations: Broadcasting, Reshaping and Einsum
03
Machine Learning
Regression, classification, trees, ensembles, clustering, evaluation and learning theory.
- 01What Is Machine Learning? The Learning Problem Formalised
- 02Types of Machine Learning: Supervised, Unsupervised, Self-Supervised and Reinforcement
- 03The Machine Learning Workflow: From Problem to Deployed Model
- 04Linear Regression from First Principles
- 05Polynomial Regression and Basis Functions: Non-Linearity with Linear Models
- 06Regularisation: Ridge, Lasso and Elastic Net
- 07Logistic Regression: Probabilistic Classification Done Right
- 08Softmax Regression and Multiclass Classification Strategies
- 09The Bias–Variance Trade-off: Derivation and Intuition
- 10Overfitting and Underfitting: Diagnosis with Learning Curves
- 11Train, Validation and Test Splits — and Cross-Validation Done Right
- 12Evaluation Metrics for Classification: Accuracy, Precision, Recall and F1
- 13ROC Curves, AUC and Precision–Recall Curves
- 14Regression Metrics: MSE, RMSE, MAE, R² and Beyond
- 15k-Nearest Neighbours: Learning by Similarity
- 16Naive Bayes Classifiers: Simple, Fast and Surprisingly Strong
- 17Decision Trees: Splitting Criteria, Pruning and Interpretability
- 18Bagging and the Bootstrap: Variance Reduction by Averaging
- 19Random Forests: Decorrelated Trees and Robust Predictions
- 20Boosting I: AdaBoost and the Power of Weak Learners
- 21Boosting II: Gradient Boosting Machines
- 22XGBoost, LightGBM and CatBoost: Modern Gradient Boosting Libraries
- 23Support Vector Machines: Maximum-Margin Classification
- 24Kernel Methods and the Kernel Trick
- 25k-Means Clustering: Algorithm, Objective and Pitfalls
- 26Hierarchical Clustering and Dendrograms
- 27DBSCAN and Density-Based Clustering
- 28Gaussian Mixture Models and the EM Algorithm
- 29Principal Component Analysis (PCA): Theory and Practice
- 30t-SNE and UMAP: Visualising High-Dimensional Data
- 31Linear Discriminant Analysis: Supervised Dimensionality Reduction
- 32Feature Engineering: Turning Raw Data into Signal
- 33Feature Scaling and Normalisation: Standardisation, Min–Max and Robust Scaling
- 34Handling Missing Data: Mechanisms, Imputation and Indicators
- 35Encoding Categorical Variables: One-Hot, Ordinal, Target and Beyond
- 36Learning from Imbalanced Data
- 37Feature Selection: Filter, Wrapper and Embedded Methods
- 38Hyperparameter Tuning: Grid, Random, Bayesian and Early-Stopping Methods
- 39Anomaly Detection: Finding the Unusual
- 40Ensemble Learning III: Voting, Stacking and Blending
- 41Recommender Systems I: Content-Based and Collaborative Filtering
- 42Recommender Systems II: Matrix Factorisation and Latent Factors
- 43Time Series Forecasting: Stationarity, ARIMA and Evaluation
- 44Semi-Supervised Learning: Learning from Few Labels and Many Unlabelled Examples
- 45Active Learning: Letting the Model Choose What to Label
- 46Learning Theory: PAC Learning, VC Dimension and Generalisation Bounds
- 47Association Rule Mining: Apriori, FP-Growth and Market-Basket Analysis
04
Deep Learning
Neural networks, backpropagation, optimisers, normalisation, CNNs, RNNs and training at scale.
- 01From Biological Neurons to Artificial Neural Networks
- 02The Perceptron: The First Learning Machine
- 03Multilayer Perceptrons and the Universal Approximation Theorem
- 04Activation Functions: Sigmoid, Tanh, ReLU, GELU, SwiGLU and Softmax
- 05Loss Functions in Deep Learning: What Are We Really Optimising?
- 06Backpropagation Derived Step by Step
- 07Computational Graphs and Automatic Differentiation
- 08Optimisers I: SGD, Momentum and Nesterov Acceleration
- 09Optimisers II: AdaGrad, RMSProp, Adam and AdamW
- 10Learning Rate Schedules, Warm-up and the LR Range Test
- 11Weight Initialisation: Xavier, He and Why It Matters
- 12Vanishing and Exploding Gradients — Causes and Cures
- 13Batch Normalisation: Faster, More Stable Training
- 14Beyond BatchNorm: Layer, Group, Instance and RMS Normalisation
- 15Dropout: Regularisation by Random Deletion
- 16Regularisation in Deep Learning: Weight Decay, Early Stopping, Augmentation and More
- 17Convolutional Neural Networks: The Core Ideas
- 18Padding, Stride, Pooling and Receptive Fields
- 19Recurrent Neural Networks: Modelling Sequences
- 20Long Short-Term Memory (LSTM): Gated Memory Explained
- 21Gated Recurrent Units (GRU) and Choosing a Recurrent Cell
- 22Sequence-to-Sequence Models and the Encoder–Decoder Framework
- 23The Attention Mechanism: Learning Where to Look
- 24Residual Connections: Why Very Deep Networks Became Trainable
- 25Autoencoders: Compression, Denoising and Representation Learning
- 26Embeddings: Turning Discrete Things into Meaningful Vectors
- 27Transfer Learning and Fine-Tuning
- 28PyTorch Fundamentals: Tensors, Autograd, Modules and the Training Loop
- 29TensorFlow and Keras Fundamentals
- 30Debugging Neural Network Training: A Systematic Recipe
- 31Mixed-Precision Training and GPU Efficiency
- 32Distributed Training: Data, Model, Pipeline and Sharded Parallelism
- 33Graph Neural Networks: Learning on Relational Data
- 34Knowledge Distillation: Teaching Small Models with Large Ones
- 35Model Compression: Pruning and Quantisation
- 36Neural Architecture Search and Automated Machine Learning
- 37Double Descent and the Generalisation Mystery of Deep Learning
- 38Loss Landscapes, Saddle Points and Flat Minima
05
Computer Vision
From pixels to perception: classification, detection, segmentation, ViTs and 3D vision.
- 01Introduction to Computer Vision: From Pixels to Perception
- 02Image Processing Fundamentals: Filtering, Convolution and Edge Detection
- 03Classical Features: Harris Corners, SIFT, HOG and Bag of Visual Words
- 04LeNet and AlexNet: The Birth of Deep Vision
- 05VGG and GoogLeNet/Inception — Depth and Multi-Scale Design
- 06ResNet in Depth: Architecture, Bottlenecks and Variants
- 07EfficientNet and Principled Model Scaling
- 08MobileNet and Efficient Architectures for Edge Devices
- 09Building an Image Classification Pipeline End to End
- 10Data Augmentation for Computer Vision
- 11Object Detection I: R-CNN, Fast R-CNN and Faster R-CNN
- 12Object Detection II: YOLO and Real-Time Detection
- 13Object Detection III: SSD, RetinaNet and the Focal Loss
- 14Semantic Segmentation: FCN, U-Net and DeepLab
- 15Instance Segmentation: Mask R-CNN and Beyond
- 16Vision Transformers (ViT): Images as Sequences of Patches
- 17Self-Supervised Vision: SimCLR, MoCo, DINO and Masked Autoencoders
- 18CLIP: Connecting Images and Language
- 19Human Pose Estimation
- 20Face Recognition: Metric Learning, ArcFace and Responsible Use
- 21Video Understanding: Action Recognition and Temporal Modelling
- 223-D Vision: Stereo, Depth Estimation, Point Clouds and NeRF
- 23OCR and Document AI: From Scanned Forms to Structured Data
- 24AI in Medical Imaging: Opportunities, Pitfalls and Validation
- 25Vision Foundation Models: Segment Anything and Promptable Vision
- 26Explaining Vision Models: Saliency Maps, Grad-CAM and Their Limits
- 27Adversarial Examples: Fooling Neural Networks and Defending Them
06
NLP & Transformers
Language models, embeddings, attention, BERT, GPT, speech and multilingual NLP.
- 01Introduction to Natural Language Processing
- 02Text Preprocessing: Tokenisation, Normalisation, Stemming and Lemmatisation
- 03Bag of Words and TF-IDF: Classical Text Representation
- 04N-gram Language Models, Smoothing and Perplexity
- 05Word2Vec: Learning Word Embeddings from Context
- 06GloVe and FastText: Global Statistics and Subword Embeddings
- 07Subword Tokenisation: BPE, WordPiece, Unigram and SentencePiece
- 08Text Classification: From Linear Models to Fine-Tuned Transformers
- 09Sentiment Analysis: Opinions, Aspects and Nuance
- 10Named Entity Recognition: Finding People, Places and Organisations
- 11Part-of-Speech Tagging and Conditional Random Fields
- 12Neural Machine Translation: From Seq2Seq to Multilingual Transformers
- 13The Transformer Architecture Explained, Block by Block
- 14Self-Attention in Depth: Intuition, Complexity and Variants
- 15Positional Encodings: Sinusoidal, Learned, RoPE and ALiBi
- 16BERT: Bidirectional Encoder Representations from Transformers
- 17The GPT Family: Autoregressive Language Models from GPT-1 to Today
- 18T5 and BART: Encoder–Decoder Pretraining and Text-to-Text Learning
- 19Fine-Tuning Pretrained Language Models: A Practical Guide
- 20Question Answering: Extractive, Open-Domain and Generative
- 21Text Summarisation: Extractive and Abstractive Methods
- 22Evaluating NLP Systems: Perplexity, BLEU, ROUGE, BERTScore and Human Judgement
- 23Information Retrieval and Semantic Search
- 24Topic Modelling: LDA, NMF and Neural Topic Models
- 25Multilingual and Low-Resource NLP (with a Focus on Bangla)
- 26Dialogue Systems and Chatbots: From Rules to LLM Assistants
- 27Automatic Speech Recognition: From HMMs to Whisper
- 28Text-to-Speech: From Concatenation to Neural Voices
- 29Efficient Transformers: Sparse Attention, Linear Attention and FlashAttention
07
Generative AI & LLMs
VAEs, GANs, diffusion, large language models, RAG, fine-tuning and AI agents.
- 01Generative vs Discriminative Models: Learning to Create
- 02Variational Autoencoders: Probabilistic Latent Spaces
- 03Generative Adversarial Networks: The Generator–Discriminator Game
- 04GAN Variants: Conditional GANs, Pix2Pix, CycleGAN and StyleGAN
- 05Normalising Flows: Exact Likelihood with Invertible Networks
- 06Diffusion Models: Generating by Learning to Denoise
- 07Latent Diffusion and Stable Diffusion: Text-to-Image at Scale
- 08Guidance in Diffusion Models: Classifier and Classifier-Free Guidance
- 09Large Language Models: What They Are and How They Are Built
- 10Scaling Laws: How Performance Grows with Compute, Data and Parameters
- 11Pretraining LLMs: Data Pipelines, Objectives and Infrastructure
- 12Instruction Tuning: Teaching Language Models to Follow Directions
- 13RLHF: Reinforcement Learning from Human Feedback
- 14Direct Preference Optimisation (DPO) and Beyond
- 15Prompt Engineering: Getting Reliable Results from LLMs
- 16Chain-of-Thought and Reasoning in Language Models
- 17In-Context Learning: How LLMs Learn from Prompts
- 18Retrieval-Augmented Generation (RAG): Grounding LLMs in Your Documents
- 19Vector Databases and Approximate Nearest Neighbour Search
- 20LoRA and Parameter-Efficient Fine-Tuning (PEFT)
- 21Quantising Large Language Models for Efficient Inference
- 22Mixture of Experts: Scaling Parameters Without Scaling Compute
- 23LLM Inference: KV Caching, Batching and Speculative Decoding
- 24Decoding Strategies: Greedy, Beam Search, Temperature, Top-k and Top-p
- 25Hallucination in LLMs: Causes, Detection and Mitigation
- 26Evaluating Large Language Models: Benchmarks, Arenas and Custom Evals
- 27AI Agents and Tool Use: LLMs That Act
- 28Multimodal Models: Vision–Language and Beyond
- 29Text-to-Image and Text-to-Video Generation: Systems, Control and Provenance
- 30Building an LLM Application End to End: From Idea to Production
08
Reinforcement Learning
MDPs, dynamic programming, Q-learning, policy gradients, PPO and AlphaZero.
- 01Introduction to Reinforcement Learning: Learning by Interaction
- 02Markov Decision Processes: The Mathematical Framework of RL
- 03The Bellman Equations: Recursive Structure of Value
- 04Dynamic Programming: Policy Evaluation, Policy Iteration and Value Iteration
- 05Monte Carlo Methods in Reinforcement Learning
- 06Temporal-Difference Learning: Learning from Guesses
- 07Q-Learning and SARSA: Model-Free Control
- 08Multi-Armed Bandits: The Exploration–Exploitation Dilemma
- 09Function Approximation in RL: From Tables to Neural Networks
- 10Deep Q-Networks (DQN): Human-Level Atari from Pixels
- 11Improving DQN: Double, Dueling, Prioritised Replay and Rainbow
- 12Policy Gradient Methods: REINFORCE and the Policy Gradient Theorem
- 13Actor–Critic Methods: A2C, A3C and Advantage Estimation
- 14PPO and TRPO: Stable Policy Optimisation with Trust Regions
- 15Continuous Control: DDPG, TD3 and Soft Actor–Critic
- 16Model-Based Reinforcement Learning: Learning and Planning with World Models
- 17AlphaGo, AlphaZero and MuZero: Search Meets Deep Learning
- 18Multi-Agent Reinforcement Learning: Cooperation, Competition and Equilibria
- 19Imitation Learning and Inverse Reinforcement Learning
- 20Offline Reinforcement Learning: Learning from Logged Data
- 21Reinforcement Learning in the Real World: Robotics, Sim-to-Real and Safety
09
MLOps & Engineering
Pipelines, reproducibility, serving, monitoring and shipping ML systems to production.
- 01What Is MLOps? From Notebook to Reliable Production System
- 02Data Versioning, Validation and Pipelines
- 03Experiment Tracking: Never Lose a Result Again
- 04Reproducibility in Machine Learning
- 05Model Serving: Batch, Real-Time APIs and Streaming
- 06Docker and Containers for Machine Learning
- 07Model Monitoring: Detecting Data Drift and Concept Drift
- 08CI/CD for Machine Learning: Testing and Automating ML Systems
- 09Feature Stores and Training–Serving Consistency
- 10Edge AI and TinyML: Running Models on Phones and Microcontrollers
- 11GPUs and Hardware for Machine Learning
- 12Data Labelling and Annotation: Building High-Quality Datasets
- 13A/B Testing and Online Evaluation of ML Models
- 14From Notebook to Production Code: Structuring Clean ML Projects
- 15Model Cards, Datasheets and Responsible Documentation
10
AI Ethics, Society & Careers
Fairness, explainability, privacy, safety, regulation, humanitarian AI and your career.
- 01Why AI Ethics Matters: Principles for Responsible Engineers
- 02Bias and Fairness in Machine Learning
- 03Fairness Metrics: Demographic Parity, Equalised Odds and Calibration
- 04Explainable AI: LIME, SHAP and Interpretable Models
- 05Privacy in Machine Learning and Differential Privacy
- 06Federated Learning: Training Without Centralising Data
- 07AI Safety and Alignment: Making Capable Systems Do What We Intend
- 08AI Security: Data Poisoning, Prompt Injection, Model Theft and Defences
- 09AI Regulation and Governance: The EU AI Act and Beyond
- 10AI for Humanitarian Action and Social Good
- 11Deepfakes, Synthetic Media and Misinformation
- 12The Environmental Cost of AI: Energy, Carbon and Water
- 13AI and the Future of Work
- 14Responsible AI in Education: Learning With, Not Instead Of, AI
- 15Careers in AI: A Roadmap for Students
- 16How to Read a Machine Learning Research Paper
- 17Doing Machine Learning Research: A Guide for Your Thesis