Artificial Intelligence & Data Science is coming soon
We're still writing and reviewing the notes for every subject, and we'll open it when every subject is ready. Meanwhile you can practise aptitude, ask Aria about any subject, or see the branches that are ready.
Artificial Intelligence & Data Science
Machine Learning
Feature engineering and PCA, regression, classification and SVM, clustering and ensembles, Q-learning.
Your progress in this subject0 of 8 · 0%
- 1Types of learningOverview of the main types of machine learning (supervised, unsupervised, semi‑supervised, reinforcement, self‑supervised, transfer, online/batch, active and federated), with formal definitions, core formulas (empirical/expected risk), exam-style derivations, diagrams to draw, worked model-answer ex
- 2Models of machine learningOverview of common families of machine-learning models, how they represent functions, their assumptions, and how to choose and evaluate them.
- 3Feature engineeringFeature engineering means converting raw data into informative features (numeric or categorical) that help machine-learning models learn better; it covers scaling, encoding, missing-value handling, transformations, interaction and polynomial features, binning, and leakage avoidance.
- 4Tree-based and support vector regressionClear, exam-oriented notes on decision-tree (CART) regression and support vector regression (SVR): how they work, key formulas, derivations, worked examples, common mistakes and exam tips.
- 5K-nearest neighboursK-nearest neighbours (KNN) is a simple, instance-based supervised learning algorithm used for classification and regression that predicts a query point's label from the labels of its nearest training examples.
- 6Bagging and random forestsBagging (bootstrap aggregating) and Random Forests are ensemble methods that reduce model variance by training many decision trees on bootstrap samples and aggregating their predictions; random forests further decorrelate trees by selecting random subsets of features at each split.
- 7Boosting and stackingBoosting builds a strong predictor by adding many weak learners in a stagewise way (AdaBoost, gradient boosting); stacking combines different trained models by learning a meta-learner on their predictions.
- 8Markov decision processesMarkov decision processes (MDPs) formalise sequential decision making under uncertainty using states, actions, transition probabilities and rewards; they underpin dynamic programming and reinforcement learning methods such as value iteration and policy iteration.
Coming next in Machine Learning
- What machine learning is· notes coming
- Principal component analysis· notes coming
- Linear discriminant analysis· notes coming
- Simple and multiple linear regression· notes coming
- Polynomial regression· notes coming
- Bias-variance trade-off· notes coming
- Ridge and lasso regularisation· notes coming
- Regression metrics· notes coming
- Binary and multiclass classification· notes coming
- Classification metrics· notes coming
- Multiclass strategies and averaging· notes coming
- Support vector machines· notes coming
- Kernel functions· notes coming
- Hierarchical clustering· notes coming
- K-means and choosing k· notes coming
- DBSCAN· notes coming
- Gaussian mixture models· notes coming
- Elements of reinforcement learning· notes coming
- Bellman equation· notes coming
- Exploration and exploitation· notes coming
- Q-learning· notes coming
- Elastic net and Bayesian linear regression· notes coming
- Imbalanced classification· notes coming
- K-medoids and K-prototype· notes coming
- Voting ensembles· notes coming
- XGBoost· notes coming
- Hypothesis space and inductive bias· notes coming
- Cross-validation and model evaluation· notes coming
- Decision tree learning and overfitting· notes coming
- Collaborative filtering for recommendation· notes coming
- Bayesian learning and naive Bayes· notes coming
- Logistic regression· notes coming
- Perceptron and multilayer networks· notes coming
- Backpropagation in neural networks· notes coming
- Computational learning theory and PAC learning· notes coming
- Sample complexity and VC dimension· notes coming
- Loss functions in machine learning· notes coming
- Feature selection: filter, wrapper and embedded methods· notes coming
- ICA, CCA and NMF· notes coming
- Self-supervised learning and AutoML· notes coming
- Explainable AI and federated learning· notes coming
- Error-correcting output codes· notes coming
- KD-trees and efficient nearest-neighbour search· notes coming
- Eligibility traces and TD(λ)· notes coming
- Hopfield networks and bidirectional associative memory· notes coming
- Adaptive resonance theory: ART1 and ART2· notes coming