MGT-448: Statistical inference and machine learning

About
Privacy
Disclaimer

Graph Chatbot

Lectures in this course (47)

Kernel Methods: Feature Mapping and Kernel Matrix

Explores feature mapping in Kernel Methods and the properties of the Kernel matrix.

Support Vector Machines: Basics and Lagrange Duality

Covers the basics of Support Vector Machines and Lagrange Duality.

Support Vector Machines: Theory and Optimization

Delves into the theory and optimization of Support Vector Machines, including Mercer's Theorem and Lagrange Duality.

Optimization Framework: Inequality Constraints

Covers an optimization framework with inequality constraints and Karush-Kahn Tucker conditions.

Generative Models: Logistic Regression & Gaussian Distribution

Explores generative models, logistic regression, and Gaussian distribution for approximating posterior probabilities and optimizing model performance.

Statistical Inference and Machine Learning

Covers statistical inference, machine learning, SVMs for spam classification, email preprocessing, and feature extraction.

Building a Decision Tree

Covers building decision trees to classify mushrooms as poisonous or not.

Supervised Learning: Decision Trees

Covers supervised learning with decision trees and feature selection for classification.

Decision Trees: Overfitting and Randomization

Explores decision trees, overfitting, and randomization in supervised learning, emphasizing the importance of managing variance and feature selection.

Addressing Overfitting in Decision Trees

Explores overfitting in decision trees and introduces random forests as a solution.

Bayesian Statistics: Regularization and Divergence

Covers Kullback-Leibler divergence, regularization, and Bayesian statistics to combat overfitting in machine learning models.

Hoeffding's Inequality: Binomial Distribution

Explores Hoeffding's inequality and the binomial distribution, focusing on error minimization and generalization gaps in predictor selection.

Bayesian Parameter Estimation

Explores Bayesian Parameter Estimation, emphasizing cost minimization and reward maximization.

Bayesian Parameter Estimation

Covers an example of Bayesian parameter estimation and the trade-off between bias and variance in supervised learning.

K-Means Clustering: Image Compression

Covers K-means algorithm for image compression and PCA for dimensionality reduction.

Eigenvalues and EM Algorithm

Covers eigenvectors, principal components, likelihood variables, EM algorithm, Jensen's inequality, and maximizing lower bounds.

Principal Component Analysis: Covariance Matrix and Eigenvalues

Explores Principal Component Analysis focusing on maximizing variance and finding eigenvalues.

Latent Variable Models

Explores latent variable models, EM algorithm, and Jensen's inequality in statistical modeling.

EM Algorithm: Lite-bulbs Lifetime Estimation

Explains the EM algorithm for estimating lite-bulbs lifetime and discusses its application to heavy-bulbs.

Variational Inference: Lower Bound and ELBO

Explains variational inference, Jensen's inequality, E-step, M-step, and MCMC sampling.

Page 2 of 3