Udemy
    •  
    •  
    •  
    •  
    •  
    •  
    •  
    •  
Turn what you know into an opportunity and reach millions around the world.
Learn More
Your cart is empty.
Keep shopping
Fundamentals of Machine Learning
Rating: 4.5 out of 5(16 ratings)
2,310 students

Fundamentals of Machine Learning

This course will start your career in data science.
Created byYiqiao Yin
Last updated 6/2022
English
English [Auto],

What you'll learn

  • Learn about the fundamental principles of machine learning
  • Build customized models to use for different data science projects
  • Build customized Deep Learning models to start your own data science career
  • Start your data science career and connect with the tutor in industry

Course content

3 sections • 25 lectures • 8h 40m total length

  • Welcome1:28

    Discover the fundamentals of machine learning from statistical learning through deep learning, covering bias-variance trade-offs with sampling bootstrap, model selection, radicalization, tree-based methods, SVMs, unsupervised learning, and classification matrices.

  • Introduction8:57

    Explore the history from linear regression, discriminant analysis (lda), and logistic regression to generalized linear models and neural networks, with python and r for beginners.

  • Basics in Statistical Learning42:37
  • Linear Regression38:59

    Explore linear regression as the foundational supervised learning model, fitting sales from advertising budgets (TV, radio, newspaper) using least squares, and assess relationships with confidence intervals, hypothesis tests, and F-tests.

  • Classification22:35

    Explore how classification builds on linear regression through logistic regression, using the sigmoid link to map inputs to probabilities. Learn maximum likelihood, odds ratios, and two-class and multi-class logistic regression.

  • Q1
  • Sampling and Bootstrap14:57
  • Model Selection34:35

    Discover how to select predictive models to boost accuracy and reduce complexity. Learn best subset, forward and backward stepwise selection, and criteria such as cp, aic, eic, and adjusted r-squared.

  • Going Beyond Linearity8:38

    Go beyond linearity by transforming linear regression with polynomial features and step functions, using model selection and Cp or adjusted R-squared to choose the degree and build feature engineering indicators.

  • Tree-based Methods - Part 136:46

    Explore decision trees and regression trees that segment predictor space into regions using thresholds, and learn how random forests and BART use bagging and pruning to boost accuracy.

  • Tree-based Methods - Part 237:47
  • Q2
  • SVM22:05
  • Deep Learning57:59

    Explore deep learning foundations with artificial neural networks, convolutional neural networks, and recurrent neural networks, including key training concepts like backpropagation, cross-entropy, and data augmentation.

  • Q3
  • Unsupervised Learning51:42
  • Classification Metrics22:49

    Analyze classification metrics for model evaluation, including accuracy, confusion matrix, specificity, sensitivity, ROC AUC, and the F1 score derived from precision and recall.

  • Q4

Requirements

  • No prior mathematical or programming knowledge required. Some python programming experience is helpful.

Description

This is an introduction course of machine learning. The course will cover a wide range of topics to teach you step by step from handling a dataset to model delivery. The course assumes no prior knowledge of the students. However, some prior training in python programming and some basic calculus knowledge is definitely helpful for the course. The expectation is to provide you the same knowledge and training as that is provided in an intro Machine Learning or Artificial Intelligence course at a credited undergraduate university computer science program.


The course is comparable to the Introduction of Statistical Learning, which is the intro course to machine learning written by none other than the greatest of all: Trevor Hastie and Rob Tibshirani! The course was modeled from the "Introduction to Statistical Learning" from Stanford University.


The course is taught by Yiqiao Yin, and the course materials are provided by a team of amazing instructors with 5+ years of industry experience. All instructors come from Ivy League background and everyone is eager to share with you what they know about the industry.


The course has the following topics:

  • Introduction

  • Basics in Statistical Learning

  • Linear Regression

  • Clasification

  • Sampling and Bootstrap

  • Model Selection & Regularization

  • Going Beyond Linearity

  • Tree-based Method

  • Support Vector Machine

  • Deep Learning

  • Unsupervised Learning

  • Classification Metrics

The course is composed of 3 sections:

  1. Lecture series <= Each chapter has its designated lecture(s). The lecture walks through the technical component of a model to prepare students with the mathematical background.

  2. Lab sessions <= Each lab session covers one single topic. The lab session is complementary to a chapter as well as a lecture video.

  3. Python notebooks <= This course provides students with downloadable python notebooks to ensure the students are equipped with the technical knowledge and can deploy projects on their own.

Who this course is for:

  • Beginners in python programming, machine learning, and data science.

Report abuse