Columbia Engineering Exec Ed Applied Machine Learning | Online Certificate Program

Get Your Brochure

First name

Last Name

Email

Country

Work Experience

Phone (optional)

By clicking the button below, you agree to receive communications from Columbia Engineering & Emeritus about this program and other relevant programs. Privacy Policy (optional)

STARTS ON

May 25, 2021

DURATION

5 months, online
8-10 hours per week

PROGRAM FEE

US$2,350

Flexible payment available

Special group enrollment pricing

Who is this course for?

The Applied Machine Learning course teaches you a wide-ranging set of techniques of supervised and unsupervised machine learning approaches using Python as the programming language.

Since this course requires an intermediate knowledge of Python, you will spend the first part of this course learning Python for Data Analytics taught by Emeritus. This will provide you with the programming knowledge required to do the assignments and application projects that are part of the Applied Machine Learning course.

If you are looking to implement or lead a machine learning project or looking to incorporate machine learning capability in your software application, this course is appropriate for you. This is a programming course: you will be required to write code, but no prior programming knowledge is required.

PREREQUISITES:

The course requires an undergraduate knowledge of statistics (descriptive statistics, regression, sampling distributions, hypothesis testing, interval estimation etc.), calculus (derivatives), linear algebra (vectors & matrix transformation) and probability (conditional probability/Bayes theorem).

*Assessment: Students will be given an assessment to test their math skills prior to commencement of the course. You can view sample questions by clicking here. To familiarize yourself with the topics of the assessment, refer to learning resources by clicking here.

$57.6B

anticipated spending growth on AI and ML by 2021.

SOURCE: INTERNATIONAL DATA CORPORATION (IDC)

16%

expected wage growth for data scientists (vs. <2% average wage increase across all occupations).

SOURCE: U.S. BUREAU OF LABOR STATISTICS

225%

decrease in ‘click-to-ship’ time by Amazon using ML algorithm.

SOURCE: FORBES

Download Brochure

Course Highlights

Decorative image relating to 240+ Faculty Video Lectures

Decorative image relating to 45 Quizzes / Assignments

Decorative image relating to 18 Moderated Discussion Boards

Decorative image relating to 20+ Q&A Sessions with Course Leaders

Decorative image relating to 12 Application Projects

Decorative image relating to Includes Live Online Teaching

Learning Journey

Going beyond the theory, our approach invites participants into a conversation, where learning is facilitated by live subject matter experts and enriched by practitioners in the field of machine learning:

Define a model for your data and make the model learn.
Build regression models to predict an unknown output from a given set of inputs.
Create classification models to categorize datasets such as email messages as spam or non-spam.
Develop unsupervised models like topic models or recommender systems to extract hidden patterns from large amounts of data
Determine hidden parameters in data to improve the accuracy of your model's predictions.
Create probabilistic data models to predict a range of possible outcomes that account for real-world risks and uncertainties.

Program Topics

Designed to teach you models and methods used in machine learning for real-world applications such as recommender systems and classification models, this 5-month program begins by building your foundational skills in Python, followed by the supervised and unsupervised learning techniques of applied machine learning.

Part 1: Python for Data Analytics (Video content and delivery by Emeritus)

Module 1: Introduction to Data Science
Learn the tools, skills, and common workflows of data scientists, identify bias in data science, and understand reproducibility, collaboration, and communication.
Module 2: Working with Data Types and Operators in Python
Learn to code in Jupyter Notebooks and use various data types in Python including strings, integers, floats, Booleans, lists, and tuples.
Module 3: Writing Functions in Python
Use functions and methods to perform analyses and identify correct syntax of functions along with uses of if statements and loops.
Module 4: Popular Data Science Packages in Python
Perform basic analyses and define proper syntax using NumPy and Pandas, import and use packages in Jupyter Notebooks, and filter and slice data frames with Pandas.
Module 5: Advanced Functions
Analyze data frames with lambda functions and construct complex functions with multiple parameters, nested functions, as well as functions with default arguments;
Module 6: Data Manipulation and Analysis with Pandas
Use Pandas to build, extract, filter, and transform data frames as well as to manipulate and transform a dataset from Kaggle.

Module 7: Data Visualization with Matplotlib
Draw insights from your data and communicate recommendations to others by creating data visualizations such as basic plots, source codes, and time series using Matplotlib.
Module 8: Random Variables and Statistical Inferences
Examine probability as well as statistics in Python, differentiate between standard deviation and variance in a dataset, and collaborate on GitHub.
Module 9: Statistical Distributions and Hypothesis Testing
Use normal distribution, t-distribution, and Bernoulli distribution in Python, test a model using P values as well as confidence intervals, and conduct basic hypothesis testing.
Module 10: Data Cleaning
Examine and clean different types of data across data types to prepare it for analysis and place data into a Tidy Data format.
Module 11: Exploratory Data Analysis
Learn new methods to explore and visualize your data with Matplotlib, define key characteristics of exploratory data analysis methods, and use them to describe data.
Module 12: Getting Started with Linear Algebra for Machine Learning
Apply foundational linear algebra concepts in Python to build basic algorithms, solve equations with matrix operations, and perform advanced linear algebra procedures.

Part 2: Applied Machine Learning (Video content from Columbia Engineering and delivery by Emeritus)

Supervised Learning

Module 1: Regression
Perform supervised learning with probabilistic models, linear regression, and maximum likelihood, use matrix operations to code least squares, and implement Ordinary Least Squares.
Module 2: Ridge Regression
Analyze the bias-variance tradeoff through least squares and ridge regression, apply the Bayes rule to quantify uncertainty, and use regression analysis to solve a real-world problem.
Module 3: Bayesian Methods
Understand “active learning,” the Lagrange multipliers tool, and the sparse regression model, identify MAP solutions, estimate covariants, and apply Bayesian linear regression.
Module 4: Classification Algorithms
Deal with classification using nearest neighbors, K-nearest neighbors, and the Bayes classifier, predict outcomes with supervised learning classification, and use the perceptron algorithm.
Module 5: Logistic Regression, Kernel Methods, and Gaussian Processes
Analyze logistic regression and its algorithm, learn about a kernel, kernelized perceptron, the Nadaraya-Watson regression model, and Gaussian processes.
Module 6: Support Vector Machines and Decision Trees
Plot hyperplanes with the maximum margin method, modify SVM, code decision tree-based classifiers to make predictions about variables, and organize grid search hyperparameters.
Module 7: Boosting and K-Means Clustering
Understand a bootstrap dataset and decision stump, learn about bagging and boosting techniques, identify characteristics of K-means tools, and use a label encoder.

Unsupervised Learning

Module 8: Clustering Methods
Code functions in Python implementing K-means, soft K-means, and Gaussian mixture model clustering to assign points and recalculate clusters.
Module 9: Recommendation Systems
Employ recommendation systems to predict a user score, compare LDA vs. PMF vs. NMF, translate a mathematical algorithm into code, and use SVD to make item recommendations.
Module 10: Principal Component Analysis and Markov Models
Use PCA to code a recommender system in Python, implement PCA using sklearn, prepare and scale data in preparation for PCA, and implement Markov chains using quantecon.
Module 11: Hidden Markov Models and Kalman Filtering
Make predictions about variables using the Hidden Markov Model, differentiate between PCAs, Markov models, and Gaussian models, and use forward/backward algorithm to solve HMM.
Module 12: Association Analysis and Model Selection
Apply association analysis to a dataset, model future stock prices using sequential methods, apply one-hot encoding to a data frame, and use the library Mlxtend for association analysis.

Download Brochure

Application Projects

Decorative image relating to Movie Recommendation Engine

Movie Recommendation Engine

You will build a movie recommendation engine by applying collaborative filtering and topic modelling techniques. You use a dataset which contains 20 million viewer ratings of 27,000 movies.

Decorative image relating to House Price Prediction

House Price Prediction

You will write code to predict house prices based on several parameters available in the Ames City dataset compiled by Dean De Cock using least squares linear regression and Bayesian linear regression.

Decorative image relating to Human Activity Prediction

Human Activity Prediction

You will predict the human activity (walking, sitting, standing) that corresponds to the accelerometer and gyroscope measurements by applying the nearest neighbours technique.

Decorative image relating to Credit Card Fraud Detection

Credit Card Fraud Detection

You will detect potential frauds using credit card transaction data. You will apply the random forest method to identify fraudulent transactions.

Decorative image relating to Marketing Segmentation

Marketing Segmentation

You will create market segments using the US Census dataset and by applying the k-means clustering method.

Load more

Note: All product and company names are trademarks™ or registered® trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them.

Faculty

Dr. John W. Paisley

Dr. John W. Paisley

Columbia University Associate Professor, Electrical Engineering Affiliated Member, Data Sciences Institute

John has a PhD from Duke and has been a postdoctoral researcher in the Computer Science departments at Princeton University and UC Berkeley. John Paisley’s research focuses on developing models for large-scale text and image processing applications. He is particularly interested in Bayesian models and posterior inference techniques that address the big data problem.

Course FAQs

What is the process to enroll for the course?

1. Review the sample math assessment, to see if you feel confident with the material that includes Calculus, Linear Algebra, Statistics, and Probability.

2. Review the learning resources to familiarize yourself with the assessment quiz content.

3. Submit your application.

4. Take the math assessment. You will have two attempts. Participants who pass the assessment will receive access to the course. Any deposit fees will be refunded to participants who do not pass the assessment.
Why should a Statistician/Mathematician enroll in this course?

The course familiarizes you with Machine learning algorithms and applications. It will also help you understand the approach to a business problem and provide you with the tool knowledge needed to transition to a Machine Learning or a Data Science role.
Why should a Data Scientist/Data Engineer/Big Data Engineer enroll in this course?

The course familiarizes you with Machine learning algorithms and applications and provides a solid foundation in statistics/mathematics and problem-solving skills to help you solve enterprise-level problems. The Applied Machine Learning course augments your existing knowledge of various tools and expands your skill set as a Data Science or Machine Learning professional.
Why should a Software Developer enroll in this course?

The course familiarizes you with Machine learning algorithms and applications while providing a solid foundation in statistics/mathematics and enhancing your business acumen. It augments your existing programming knowledge and expands the technologies you are familiar with, helping you further develop your skill set as a Data Science or Machine Learning professional.
My current role in my company does not include exposure to Machine Learning. Does it make sense for me to enroll in this course?

Absolutely! Knowledge of Data Science and Machine Learning (ML) has quickly become a requisite across industries, and all businesses will eventually need to use these techniques to thrive. While your current role may not require Machine learning knowledge, it is almost certain that ML skills will be in high demand in most every industry in the future.
How can I implement the concepts and ideas discussed in the course?

The course is a blend of theory, tools, and case studies (datasets) that are easy to assimilate and implement. For instance, students work on application projects that require them to apply the Machine Learning concepts they’ve learned to datasets and derive inferences. These application projects are intentionally made to be challenging, and students are expected to spend substantial time and effort solving them. At the end of the course, students will be able to apply Machine Learning to solve many of the business problems they face in their workplace.
What is the relationship between Emeritus and Columbia Engineering Executive Education?

Columbia Engineering Executive Education is collaborating with online education provider Emeritus to offer a portfolio of high-impact online courses. These courses leverage Columbia’s thought leadership in management practice developed over years of research, teaching, and practice.
What software and versions will I need in this course?
Recommended System Requirements
- Processors: 2.60 GHz
- RAM: 8 GB of RAM
- Disk space: 2 to 3 GB
- Operating systems: Windows 10, MacOS and Linux
- Python download link
- Compatible tools: Any text editor, Command prompt
Minimum System Requirements
- Processors: 1 GHz
- RAM: 1 GB of RAM
- Disk space: 1 GB
- Operating systems: Windows 7 or later, MacOS and Linux
- Python versions: 2.7.X, 3.6.X
- Compatible tools: Any text editor, Command prompt

Certificate

Certificate

Upon successful completion of the course, participants will receive a verified digital certificate from Emeritus in collaboration with Columbia Engineering Executive Education.

Download Brochure

Apply Now

Early registrations are encouraged. Seats fill up quickly!

Flexible payment options available. Learn more.