COMP3308 HD Summary (90/100) - Complete AI & Machine Learning Revision Guide

Description

Achieved a High Distinction (90/100) using these meticulously typed, highly structured summary notes. These notes condense complex AI algorithms and heavy mathematics into easy-to-understand, bite-sized rules perfect for exam revision. Topics Covered: - Intro to AI: Rational Agents & Turing Tests - Search Strategies: Uninformed (BFS, DFS, UCS) & Informed (A*, Heuristics) - Local Search & Optimisation: Hill-Climbing, Beam Search, Simulated Annealing, Genetic Algorithms - Adversarial Search (Game Playing): Minimax & Alpha-Beta Pruning, Expectminimax - Machine Learning Basics: Supervised vs Unsupervised, k-NN & Distance Measures, 1R, Naive Bayes - Evaluating Classifiers: Confusion Matrices, Precision/Recall/F1, K-fold Cross-Validation - Decision Trees: ID3 Algorithm, Entropy, Information Gain, Overfitting & Pruning - Neural Networks: Perceptrons, MLPs, Backpropagation & Gradient Descent - Deep Learning: CNNs (Local Connectivity, Pooling) & Autoencoders - Support Vector Machines (SVM): Maximum Margin, Hard/Soft Margins & The Kernel Trick - Ensembles of Classifiers: Bagging, Boosting (AdaBoost) & Random Forests - Probabilistic Reasoning: Bayes Theorem, Conditional Independence & Bayesian Networks (Variable Elimination/Inference) - Unsupervised Learning: k-Means Clustering, Hierarchical Clustering & Davies-Bouldin Index