Course

CS-233(b): Introduction to machine learning (BA4)

Lectures in this course (21)

Introduces Multi-Layer Perceptrons (MLPs) and covers logistic regression, reformulation, gradient descent, AdaBoost, and practical applications.

Shape Design: Optimization and Simulation

Explores shape design, optimization, and simulation using geodesic CNNs for aerodynamic performance.

Decision Forests: Structure and Training

Covers decision forests, training, weak learners, entropy, boosting, 3D pose estimation, and practical applications.

Machine Learning Basics

Introduces the basics of machine learning, covering supervised classification, decision boundaries, and polynomial curve fitting.

KNN: Nearest Neighbor Classifier

Covers the k-Nearest-Neighbor classifier, hand-written digit recognition, multi-class k-NN, data reduction, applications, graph construction, limitations, and the curse of dimensionality.

Python/NumPy Primer

Introduces Python basics and NumPy for scientific computing, covering data types, functions, arrays, indexing, and common operations.

Linear Classification: Parameterizing Lines and Distance Calculation

Covers the parameterization of lines in 2D space and the perceptron algorithm.

K-Means Clustering: Basics and Applications

Introduces K-Means Clustering, a simple yet effective algorithm for grouping data points into clusters.

Max-Margin Classifiers

Explores maximizing margins for better classification using support vector machines and the importance of choosing the right parameter.

Linear Classification: Signed Distance and Perceptron

Explores signed distance, perceptron, logistic regression, cross entropy, and multi-class classification.

Optimization Basics

Introduces optimization basics, covering logistic regression, derivatives, convex functions, gradient descent, and second-order methods.

Support Vector Machines: Non-Linear Data Mapping

Explores mapping non-linear data to higher dimensions using SVM and covers polynomial feature expansion, regularization, noise implications, and curve-fitting methods.

Adaboost: Boosting Methods

Explains Adaboost algorithm for building strong classifiers from weak ones, with a focus on boosting methods and face detection.

Kernel SVM: Polynomial Expansion & Cover's Theorem

Explores polynomial expansion, high-dimensional spaces, Cover's Theorem, Lagrangian formulation, and practical SVM applications.

Universal Approximation Theorem: MLP

Covers Multi-Layer Perceptrons (MLP) and their application from classification to regression, including the Universal Approximation Theorem and challenges with gradients.

Convolutional Neural Networks

Introduces Convolutional Neural Networks, covering fully connected layers, convolutions, pooling, PyTorch translations, and applications like hand pose estimation and tubularity estimation.

Linear Dimensionality Reduction

Explores linear dimensionality reduction through PCA, variance maximization, and real-world applications like medical data analysis.

Alpha Go: Machine Learning and Game Strategies

Explores the historical perspective and development of the AlphaGo algorithm, focusing on machine learning and game strategies.

Non-Linear Dimensionality Reduction

Covers non-linear dimensionality reduction techniques using autoencoders, deep autoencoders, and convolutional autoencoders for various applications.

Linear Dimensionality Reduction: PCA and LDA

Explores PCA and LDA for linear dimensionality reduction in data, emphasizing clustering and class separation techniques.