Designing a Machine Learning Model

By Janani Ravi
This course covers the important differences between various canonical problems in machine learning, as well as the considerations in choosing the right solution techniques, based on the specifics of the problem you are trying to solve and the data that you have available.
Play course overview
Course info
Level
Intermediate
Updated
Aug 13, 2019
Duration
3h 25m
Table of contents
Course Overview
Course Overview 2m
Exploring Approaches to Machine Learning
Module Overview 1m Prerequisites and Course Outline 2m A Case Study: Sentiment Analysis 7m Sentiment Analysis as a Binary Classification Problem 2m Rule Based vs. ML Based Analysis 7m Traditional Machine Learning Systems 4m Representation Machine Learning Systems 2m Deep Learning and Neural Networks 5m Traditional ML vs. Deep Learning 3m Traditional ML Algorithms and Neural Network Design 5m Module Summary 1m
Choosing the Right Machine Learning Problem
Module Overview 1m Choosing the Right Machine Learning Problem 7m Supervised and Unsupervised Learning 8m Reinforcement Learning 7m Recommendation Systems 4m Module Summary 2m
Choosing the Right Machine Learning Solution
Module Overview 2m Regression Models 2m Choosing Regression Algorithms 4m Evaluating Regression Models 5m Types of Classification 4m Choosing Classification Algorithms 3m Evaluating Classifiers 4m Clustering Models 5m The Curse of Dimensionality 5m Dimensionality Reduction Techniques 3m Module Summary 1m
Building Simple Machine Learning Solutions
Module Overview 1m Install and Set Up 2m Exploring the Regression Dataset 3m Simple Regression Using Analytical and Machine Learning Techniques 5m Multiple Regression Using Analytical and Machine Learning Techniques 2m Exploring the Classification Dataset 3m Classification Using Logistic Regression 4m Classification Using Decision Trees 3m Clustering Using K-means 7m Dimensionality Reduction Using Principal Component Analysis 4m Dimensionality Reduction Using Manifold Learning 5m Module Summary 1m
Designing Machine Learning Workflows
Module Overview 1m The Machine Learning Workflow 4m Case Study: PyTorch on the Cloud 6m Ensemble Learning 6m Averaging and Boosting, Voting and Stacking 2m Custom Neural Networks: Their Characteristics and Applications 4m Module Summary 1m
Building Ensemble Solutions and Neural Network Solutions
Module Overview 1m Classification Using Hard Voting and Soft Voting 5m Exploring and Preprocessing the Regression Dataset 3m Regression Using Bagging and Pasting 5m Regression Using Gradient Boosting 4m Regression Using Neural Networks 8m Summary and Further Study 2m
Description
Course info
Level
Intermediate
Updated
Aug 13, 2019
Duration
3h 25m
Description

As Machine Learning explodes in popularity, it is becoming ever more important to know precisely how to frame a machine learning model in a manner appropriate to the problem we are trying to solve, and the data that we have available. In this course, Designing a Machine Learning Model you will gain the ability to appropriately frame your use-case and then choose the right solution technique to model it. First, you will learn how rule-based systems and ML systems differ and how traditional and deep learning models work. Next, you will discover how supervised, unsupervised, and reinforcement learning techniques differ from each other. You will learn how classic supervised learning techniques such as regression and classification complement classic unsupervised techniques such as clustering and dimensionality reduction. You will then understand the assumptions and outcomes of these four classes of techniques and how solutions can be evaluated. Finally, you will round out your knowledge by designing end-to-end ML workflows for canonical ML problems, ensemble learning, and neural networks. When you’re finished with this course, you will have the skills and knowledge to identify the correct machine learning problem setup, and the appropriate solution technique for your use-case.

About the author
Janani Ravi
About the author

A problem solver at heart, Janani has a Masters degree from Stanford and worked for 7+ years at Google. She was one of the original engineers on Google Docs and holds 4 patents for its real-time collaborative editing framework.

More from the author
Building Statistical Summaries with R
Advanced
3h 3m
Jan 1, 2020
Understanding Statistical Models and Mathematical Models
Beginner
2h 35m
Dec 17, 2019
Interpreting Data Using Descriptive Statistics with Python
Intermediate
2h 21m
Nov 8, 2019
More courses by Janani Ravi
Section Introduction Transcripts
Section Introduction Transcripts

Course Overview
(Music) Hi. My name is Janani Ravi, and welcome to this course on Designing a Machine Learning Model. A little about myself: I have a master's degree in electrical engineering from Stanford and have worked at companies such as Microsoft, Google, and Flipkart. At Google, I was one of the first engineers working on real-time collaborative editing in Google Docs, and I hold four patents for its underlying technologies. I currently work on my own startup, Loonycorn, a studio for high-quality video content. As machine learning explodes in popularity, it is becoming ever more important to know precisely how to frame a machine learning model in a manner appropriate to the problem we are trying to solve and the data that we have available. In this course, you will gain the ability to appropriately frame your use case and then choose the right solution technique to model it. First, you will learn how rule-based systems and ML systems differ, and how traditional and deep learning models work. Next, you will discover how supervised, unsupervised, and reinforcement learning techniques differ from each other. You will learn how classic supervised learning techniques such as regression and classification compliment classic unsupervised techniques such as clustering and dimensionality reduction. You will then understand the assumptions and outcomes of these four classes of techniques and how solutions can be evaluated. Finally, you will round out your knowledge by designing end-to-end ML workflows, for canonical ML problems, ensemble learning, as well as neural networks. When you're finished with this course, you will have the skills and knowledge to identify the correct machine learning problem setup and the appropriate solution technique for your use case.

You're the smartest person in the room

PROVE IT

Your team won't just close the skills gap

THEY'LL HURDLE IT