Machine Learning by Columbia University and edX

Advanced Machine Learning course offered by the Columbia University. Focused on maths and theory, it delivers in-depth knowledge for students with a strong foundation within a reasonable amount of time.

Created by: John W. Paisley

Produced in 2016

What you will learn

  • Predict an output based on set of inputs with supervised learning techniques
  • Understand the difference between probabilistic and non-probabilistic modeling
  • Learn Bayes rule, Laplace approximation, and Gaussian processes
  • Apply topic modeling method for unsupervised learning
  • Advanced matrix factorization algorithm
  • Master algorithms for optimization and inference
  • Statistics and mathematics methods behind machine learning
  • Much, Much more!

Quality Score

Content Quality
Video Quality
Qualified Instructor
Course Pace
Course Depth & Coverage

Overall Score : 90 / 100

Live Chat with CourseDuck's Co-Founder for Help

Need help deciding on a machine learning course? Or looking for more detail on John W. Paisley's Machine Learning by Columbia University and edX? Feel free to chat below.
Join CourseDuck's Online Learning Discord Community

Course Description

Machine Learning is the basis for the most exciting careers in data analysis today. You'll learn the models and methods and apply them to real world situations ranging from identifying trending news topics, to building recommendation engines, ranking sports teams and plotting the path of movie zombies.Major perspectives covered include:probabilistic versus non-probabilistic modelingsupervised versus unsupervised learningTopics include: classification and regression, clustering methods, sequential models, matrix factorization, topic modeling and model selection.Methods include: linear and logistic regression, support vector machines, tree classifiers, boosting, maximum likelihood and MAP inference, EM algorithm, hidden Markov models, Kalman filters, k-means, Gaussian mixture models, among others.In the first half of the course we will cover supervised learning techniques for regression and classification. In this framework, we possess an output or response that we wish to predict based on a set of inputs. We will discuss several fundamental methods for performing this task and algorithms for their optimization. Our approach will be more practically motivated, meaning we will fully develop a mathematical understanding of the respective algorithms, but we will only briefly touch on abstract learning theory.In the second half of the course we shift to unsupervised learning techniques. In these problems the end goal less clear-cut than predicting an output based on a corresponding input. We will cover three fundamental problems of unsupervised learning: data clustering, matrix factorization, and sequential models for order-dependent data. Some applications of these models include object recommendation and topic modeling.



    • Offered by an Ivy League research university.
    • In-depth learning experience in Machine Learning.
    • Talented instructor competent in transmitting knowledge.
    • Good explanations of advanced topics such as Gaussian processes.
    • Requires strong mathematics background.
    • Lacks interactivity and practical examples.
    • Does not cover neural networks.

Instructor Details

John W. Paisley

John Paisley is an Assistant Professor in the Department of Electrical Engineering at Columbia University. John is also an affiliated member of the Data Science Institute at Columbia. John received his Ph.D. in Electrical and Computer Engineering from Duke University, where he worked with Lawrence Carin. He was then a post-doc in the Computer Science departments at Princeton University with David Blei and UC Berkeley with Michael Jordan. John's research is in the general area of statistical machine learning. His interests include probabilistic modeling and inference techniques, Bayesian nonparametric methods, dictionary learning and topic modeling.



10 total reviews

5 star 4 star 3 star 2 star 1 star
% Complete
% Complete
% Complete
% Complete
% Complete

By Kush K on 9/5/2016

There are not many courses online that provide such in-depth learning experience in Machine Learning. This course goes into some details and mathematics of the algorithms being used. It demands a good amount of time every week to understand and apply all that is being taught but that is what makes it good.

By Sedat S on 4/13/2019

Amazing in-depth knowledge. Almost all algorithms out there are being explained.But:Terrible execution, lecturer is just reading the slides and explaining a bit. Not enough real life examples or any code examples. No interactivity.

By Sensei S on 8/6/2017

This course requires a solid foundation on probabilities, calculus, linear algebra and programming. Provided these prerequisites are available (anyone who is serious about the field should possess these skills anyway), the course will become an incredibly useful resource to break into Machine Learning.

By Harish R on 10/24/2017

Beautiful course which covers advanced concepts of machine learning. Professor Paisley covers a whole range of topics and breaks down hard concepts clearly. This course is very theoretical. There are four programming assignments which give an opportunity implement some of the algorithms learned.

By Deans Charbal on 9/14/2017

This is my second course about Machine Learning (I'm a senior developer with a mathematical background) and I am delighted. This course is probably the best online ML course you can find. Professor John Paisley is an amazing and brilliant teacher ; he doesn't only have the knowledge, he has the talent of teaching, of transmitting knowledge with detail and precision, of explaining, reexplaining, illustrating sometimes complex concepts to make them simpler to understand and digest. I would like to thank him warmly, because my linear algebra and math background were pretty rusted, but it was a so-interesting 3 months, challenging for some parts but at the end of the day, it open perspectives, it provides extra tools of knowledge, it gives you keys of understanding. What a pleasure.

By Steven Frank on 5/12/2018

Buckle up -- this deep, mathematically rigorous dive into the major areas of machine learning is fast-paced and challenging. In fact, most of the course is less about machine learning than the math behind it -- problem-solving and applications take a back seat to the underlying mathematical techniques. You won't see very many implementation examples or, in the programming projects, watch a machine get smarter. For that context, you either need prior ML training or to have taken the first course in the AI MicroMaster series, Artificial Intelligence with Ansaf Salleb-Aouissi, which treats most of the topics in this course at a higher, more introductory level. Also, considerable fluency in probability and statistics is assumed -- at the level of MITx 6.041, for example. Although no textbook is suggested, I found "The Elements of Statistical Learning" by Hastie et al. to be quite useful. The topics covered are numerous, too many to list without putting you to sleep, but they span all of the common machine learning techniques except neural networks, which is a subject in itself (and is nicely covered at a high level in the AI course).

By Yogesh Luthra on 11/5/2017

This is the most insightful course I cam across. Although I am a trained practitioner of ML concepts, but there were some topics like Gaussian Processes, Collaborative Filtering, LDA etc, for which I need more satisfactory explanations.

By Henry Harya on 8/23/2017

The materials are mathematically rigorous and really provide insight on how to analyse, design and evaluate learning algorithms. Prof Paisley lectures are dense, though unfortunately he's not the greatest lecturer and spends most of the time reading from the slides.

By Kalok Kam on 5/6/2017

The course covers a wide range of machine learning techniques. The teacher is nice and his teaching is clear and clever. The programming assignments are full of fun and I have learned a lot from them. But the setting for assignments are somehow poor because most questions only accept one attempt and none answer is shown once you failed.

By Nick Burnett on 6/21/2017

Course was a great overview of the theory behind a wide range of models. The theory gives a good intuition about what the models are doing and how best to make use of them. My effectiveness has improved, even using models I thought I was very familiar with. The lecturer was excellent and did a good job of communicating difficult concepts well. The course isn't easy. You need a strong mathematical background and a decent amount of time to get through it properly.