16 New Must Watch Tutorials, Courses on Machine Learning

avcontentteam 05 Jul, 2020 • 7 min read

Introduction

Most of us fail to acknowledge that Youtube has a massive resource center of machine learning tutorials which are free to access. You no longer need to wait for launch of new MOOCs to learn a new concept. Search it on YouTube and chance are high that you’ll find it.

Last year, we published an article on top YouTube videos featuring best ever videos on neural networks, deep learning & machine learning. No doubt, videos are enriching. But, it happens that content (video, text) gets outdated overtime. Now is the time to update your knowledge.

I wrote this article to help you discover new tools, techniques, methods, practices undertaken in machine learning since last year. Always remember, the pursuit of knowledge is similar to the life of fresh water. Never stop the flow and get a perennial fresh stream of perspectives.

16 New Must Watch Tutorials, Courses on Machine Learning

 

Are these tutorials meant for me?

Depends! if you are curious and want to learn something new or enhance your skill sets.

I’ve categorized the videos in 4 principal sections, making sure that each one of you gets to learn something new. However, python users have more to learn. You’ll find full lectures, practical workshops, short talks indicating the increased dominance of machine learning in real world. Alongside, you’ll also learn how machine learning is solving real world problems at Google, Pinterest & TaxiGrab.

If you plan to watch them, make a schedule. Don’t do it all in a day. Remember, the motive is not to just watch the videos, but to understand what is being taught. It takes time and discipline. Also to help you save time, I’ve provided a short summary below every video to help you decide if that’s what you should watch!

 

Table of Contents

  1. For Newbies in Machine Learning
    • How to become a data scientist
    • Essential Data Science Skills For Every Programmer
    • Beginners Guide to Data Science Competitions
    • Machine Learning Recipes
  2. New Machine Learning Courses
    • Statistical Machine Learning
    • Machine Learning Course – University of Waterloo
    • Practical ML in Python
    • Neural Networks Course by Geoff Hinton
  3. Other Useful Talks
    • Machine Learning With Imbalanced Data sets
    • Scikit Learn
    • Advanced Techniques – Deep Learning
    • Pandas for Beginners
    • Predictive Modeling with Python
  4. Machine Learning at Organizations
    • Google
    • Pinterest
    • GrabTaxi

 

1. For Newbies in Machine Learning

How to become a Data Scientist in 6 months

Duration: 56:24 mins

In this video, Tetiana Ivanova shares her journey of becoming a data scientist in just 6 months. Participating in hackathons got her started with machine learning. If you have been wondering whether to go for analytics post graduate program or become self taught, you must watch this video. Tetiana shares her real life experience of making the career move, the hardships and truth behind the facade of a higher education.

 

Essential Data Science Skills For every Programmer

Duration: 3:23:19 hrs

Data Science has several tools used for data exploration, visualization to modeling. Which are the must to have? Andy reveals the most important tools every person should use while working on Python. These tools aren’t just easy to learn, but also upgrade your style of coding outputs. It is a must watch for python beginners. Also, he demonstrates using these tools to produce different outputs. You should install these tools as instructed and practice alongside while watching it.

 

Beginner’s Guide to Machine Learning Competitions

Duration: 1:43:08 hr

When will I win my first data science competition? I’m sure everyone asks this question to themselves. Winning world level competitions but not impossible. Little bit guidance and practice does the trick.  This tutorial trains you to solve Kaggle competition using an effective ML approach. The packages used in the tutorial are IPython notebook, scikit-learn, pandas and NLTK. You will learn about the process to follow in competitions, and how to perform modeling, feature selection, optimization & validation.

 

Machine Learning Recipes

Duration: NA

Nothing would introduce you to technical aspects of machine learning faster than these videos. Google released these 7 machine learning recipes this year. These are short tutorial (~10 mins) which cover crucial aspects of machine learning such as feature extraction, decision tree visualization, classification model, tensor flow etc. The language used is Python. However, the conceptual knowledge is tool agnostic. I think, these videos can also be watched during lunch time!

 

2. Machine Learning Courses

Statistical Machine Learning

This course was being taught at Carnegie Mellon University (CMU) in Spring 2016 session. As its name, the professor teaches topics such as regression, clustering, boosting, graphical models, minimax theory etc. This course is best suited for students having basic understanding of statistics & probability. It’s a core mathematical course, therefore you should also be comfortable with understanding mathematical equations. Alongside, there are assignments & solution which would further improve your concepts.

 

Machine Learning Course – University of Waterloo

This detailed machine learning course from University of Waterloo, will guide you through basics & advanced chapters of machine learning. It’s a conceptual course which will educate you on mathematical relations in ML algorithms. It has been taught by multiple professors including Shai Ben David, author of book Understanding machine learning. It covers topics such as linear regression, bayesian, trees, clustering, neural networks, ensemble, hidden markov model and much more. Check out the other course material here.

The first video in an introductory to the course so, feel free to skip the first 8 mins of the video.

 

Practical Machine Learning Tutorial with Python

Python has quickly gained recognition in machine learning community. With its robust libraries and actively engaging communities, students are encouraged to learn python as their core language. If you code in python, this course will help you deepen your practical ML knowledge in python. If you are following the courses, after learning theoretical concepts from previous courses, here you’ll learn how to apply them. This playlist of 57 videos covers all important ML algorithms along with detailed version of each one.

 

Neural Networks for Machine Learning

Deep Learning, a subfield of Artificial Intelligence has progressed by contributions of great minds like Geoffery Hinton. Learning from the master itself is a blessing in itself. Isn’t it? This course on neural networks was taught by him at University of Toronto. The course is designed such that it progresses through basic topics and culminates at advanced side of neural networks. It include topics such as perceptrons, back propagation, CNN, RNN, gradient descent and a lot more in detail. It’s a must watch for deep learning, neural network enthusiasts.

 

3. Other Useful Talks

Machine Learning With Imbalanced Data sets

Duration: 27:44 minutes

Classification algorithm performs poorly when the data is skewed towards one class. This problem is prominent in real world while dealing with fraud detection, cancer detection or medical diagnosis. There are couple of methods like re-sampling, one-class learning, cost-sensitive learning to address this problem. This tutorial will take you though different approaches to handle unbalanced data sets in fraud detection. Natalie also shares some practical advises which she learnt after working on numerous imbalanced problems.

 

Machine Learning Tutorial on Scikit-Learn

Duration: 3:03:54 hrs

This 3 hour tutorial touches upon the breadth of machine learning algorithms. The speaker, Sebastian Raschka, author of Python Machine learning, explains complex concepts using a beautiful mix of interactive images. Our brain is much more receptive to consume visual knowledge than textual or sound. This workshop was delivered at SciPy Conference 2016. He teaches supervised & unsupervised ML concepts supported with real life case studies. If you like this tutorial, here’s Part 2 of this series.

 

Advanced Techniques –  Deep Learning

Duration: 1:36:32 hr

In past few years, the techniques of image classification, segmentation and object detection have evolved tremendously with Deep Learning. This tutorial will take you through the advance concepts of Deep Learning focusing mainly on computer vision and image processing using Theano & Lasagne. Alongside, the speaker also discusses important tips & tricks such as dealing with less training data etc. To understand concepts, prior knowledge of algebra, calculus and machine learning is required.

 

Pandas for Beginners

Duration: 1:47:48 hr

Of all the python libraries, pandas becomes the first choice for data manipulation tasks. With its intelligent inbuilt functions, the painful task of summarizing & manipulating data becomes so much easy. This video is best suited for beginners who wants to learn python. In this tutorial, the speaker demonstrates tasks such as data selection, grouping, aggregation, plots etc. Make sure you do then side by side to develop better understanding.

 

Predictive Modeling with Python

Duration: 58:28 mins

Oliver Grisel, one of the original contributor to Scikit learn library, talks about building high performance predictive models. How do you deal with large data sets in Python? Answers to such burning questions are given in this tutorial. Alongside, he’ll also introduce you to some interesting tools which can be used in conjunction with Python to fasten our predictive modeling process. You’ll also learn about the backstage story of data and it’s chemistry with storage & distribution types.

 

4. Machine Learning at Organizations

Machine Learning : Google’s Vision

Duration: 44:44 minutes

How does google uses machine learning ? Everyone talks about it, but nobody can tell as accurately as this guy does. Learn more about google’s take on machine learning and AI, how machine learning has streamlined google’s end products. Also, it has deployed practical A.I throughout it’s products and has brought an end user more closer to the technology. Hear from Google’s machine learning leads on their breakthroughs in machine learning and their upcoming ML projects.

 

Machine Learning at Pinterest

Duration: 23:54

In this video, Jure Leskovec, Chief Scientist at Pinterest explains how machine learning is used at Pinterest. It’s motivating to see how ML is transforming ways of businesses on internet.  Here, Jure explains different segments of Pinterest driven by machine learning which affects new user experience, interest recommendation, type of content, user actions prediction, pin ranking  and visual features. Jure also shares insights about what worked for them and what lessons they learnt. I think it’s an interesting take on how machine learning is changing our day-to-day lives.

 

Machine Learning Used by Grab Taxi

Duration: 11:24

Personally to me, it is surprising to see how machine learning can solve business problems at different levels. One such example is how Grab Taxi uses machine learning to tackle the problem of taxi availability. To handle this problem, Grab started a unique initiative of bidding for a ride by the drivers and the fastest bidder wins and is assigned the ride. Watch the full video to find out how they used machine learning to build a predictive model on drivers bidding probability and used real time data to solve the problem.

 

End Notes

The tutorials listed above are meant to familiarize you with latest happenings in machine learning. Most of the videos are > 1 hour, hence you are advised to keep a schedule for watching them. Since there is an abundance of information on internet, it become crucial to find your right gems and stick to them.

These tutorials are shortlisted on the basis of upload date, view count and relevance. We made sure that none of these tutorials have got featured in our previous article. Personally, I found these videos immensely useful in demonstrating python tasks. R users might be disappointed. I tried searching for new tutorials, but couldn’t find anything helpful on youtube.

Did you like this article ? Have you seen any video yet? What about the machine learning courses? I’d love to hear your experience / suggestions in comments below.

You can test your skills and knowledge. Check out Live Competitions and compete with best Data Scientists from all over the world.

avcontentteam 05 Jul 2020

Frequently Asked Questions

Lorem ipsum dolor sit amet, consectetur adipiscing elit,

Responses From Readers

Clear

Ankur
Ankur 19 Oct, 2016

Hi I'm not of Python. Will these videos be still useful for me ?Regards, Ankur

Hossein
Hossein 19 Oct, 2016

Dear Manish the list was the best thanks for sharing

shantala
shantala 19 Oct, 2016

HI Manish,Thanks for sharing, the list was very useful....

Anon
Anon 19 Oct, 2016

Caltech's "Learning from Data", which was brought to my attention by Kunal, is back on EdX after a two-year break. Since this is conducted in parallel to the actual course at Caltech, one has the presence of mentors and the instructor himself, in the forums. Unfortunately the registration date passed just a few days ago. Of course, the separate course site is still available: http://work.caltech.edu/telecourse.html.Geoff Hinton's course (mentioned earlier) is also back in Coursera, with exercises and forums, in a do-it-at-your-pace format. https://www.coursera.org/learn/neural-networks

Jayashree
Jayashree 19 Oct, 2016

Thanks Manish for writing this airticle. Going to be immensely beneficial.

Leo
Leo 20 Oct, 2016

Very interesting topics covered. Immensely useful for someone who wants to learn Python and gain business applications of ML.

Pablov
Pablov 21 Oct, 2016

Thanks for this great selection. For newbies like me these resources are very useful! Greetings from Maldonado.

Vivek Kumar
Vivek Kumar 22 Oct, 2016

Thanks Manish!!! You posted just a awesome list of videos.

Shyamal
Shyamal 02 Nov, 2016

Hello Manish, First of all awesome work. which one of following will give me all theoretical logic about ML before moving ahead? Or should I do both? Because first course has some prerequisites. So should I skip that and will only second course give me almost all the concept to move further? 1. statistical ML (CMU) 2. Understanding ML (Waterloo)Thanks.

Mark Skrobola
Mark Skrobola 12 Nov, 2016

Great list of Videos, I created a Playlist for easy access: https://www.youtube.com/playlist?list=PLusuCu9P--ds8toAifheqXlBzjkQXEamM

Rishabh
Rishabh 22 Nov, 2016

Hi, Thanks AV Content Team, for great resources for learning ML and Python. I'm a big fan of your work. I am finding it difficult to understand statistical concepts practically. It would be of great help, if you can create similar post for statistics related courses. Thanks

  • [tta_listen_btn class="listen"]