Coursera Deep Learning Specialization

Deep Learning Specialization Review: Is It Worth Your Time?

I completed this specialisation and wrote this review to share my experience and help other learners decide whether to take it.

What is the Coursera Deep Learning Specialization?

The five-course Coursera specialisation develops neural-network foundations through Python and TensorFlow. It is a foundation for further work, not a complete education in every area of modern AI.

Time commitment: Coursera currently estimates three months at ten hours per week. My pace when I took it was faster: I could generally finish a week’s material in about half a week. That personal pace is not a promise; allow time to revisit difficult exercises.

One of the main advantages of these kind of courses is the flexibility they offer, so I would recommend you take your time to do it and try to interiorice the learning and test yourself on the content. 

Who is the Coursera Deep Learning Specialization for?

This specialization is for people that have a fair knowledge of Data Science and Machine Learning, as it explores an specific branch of Machine Learning: Deep Learning and Artificial Neural Networks. If you want to learn about general Machine Learning and take a first step into this world, there is another very famous course by Andrew Ng: The Machine Learning Specialization, which you should consider taking beforehand.

The Machine Learning Specialization can provide a useful foundation before deep learning. Completing both still needs to be followed by independent projects and practice with real data.

You can find more introductory Machine Learning courses on our Machine Learning online courses section, that will complement this Coursera Machine Learning course.

Prerequisites

Current prerequisites: the provider expects intermediate Python and recommends basic linear algebra and ML concepts. If those are gaps, use our statistics and probability course guide alongside practice with vectors, matrices and Python. The interview below provides additional background.

However, as I mentioned previously, from my point of view you will make the most out of this specialisation if you are already familiar with Machine Learning and its main concepts. 

Contents of the Coursera deep learning Specialization

Coursera Deep Learning Specialisation is composed of 5 Courses, each divided into various weeks. These courses are the following:

• Course I: Neural Networks and Deep Learning. Explains how to go from a simple neuron with a logistic regression to a full network, covering the different activation, forward and backward propagation. The most fundamental information about how ANN work is explained in this course.
• Course II: Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization. Despite not being an exact science, this course mentions how to tune the learning rate, the number of layers and neurons in each layer. Then regularisation techniques like dropout and Batch Normalization are covered, to end with an optimization section that discusses stochastic gradient descent, momentum, RMS Prop and Adam optimization algorithms. 
• Course III: Structuring Machine Learning projects. For me, this course was the most helpful of all for a Data Scientist/Engineer. It is all about how to build your projects, get results quickly and iterate to improve this results. It gives insights into how to diagnose the outcomes of our models so that we can see where the performance problem is coming from if there is one: small training set, different distributions on train and test set, over-fitting, and other problems are covered, along with their solutions. 
• Course IV: Convolutional Neural Networks. This course introduces convolutional neural networks for image tasks. It covers how CNNs work, different styles of CNNs, and also describes their origin. Then it explains the different use cases like object recognition, face Verification, face Recognition, or Neural Style Transfer. 
• Course V: Sequence models. All about Recurrent Neural Networks: how they work, why they excel in some specific tasks over traditional networks, and the different types with their pros and cons. RNNs, LSTMs, GRUs are explained along with how they are used for tasks like speech recognition, Natural Language Processing and Time Series analysis and forecast. The current syllabus also includes Hugging Face tokenizers and transformers for named entity recognition and question answering. 

Structure of the Courses

The learning format combines short videos, quizzes and notebook assignments. The exact module structure and assessment requirements can change, so use the current course dashboard for deadlines and completion rules.

The videos start with a quick introduction, progress into a series of slides covering the technical content and finish with a nice summary of what has been explained and sometimes a short notice for the following video. They are very educational and not hard at all to watch because of their duration. Some of them can be perfectly understood at video speeds greater than normal. The best video editor can effectively comprehend certain videos when played at speeds faster than normal.

In the version I took, the quizzes tested understanding of the week’s material through single-choice and multiple-choice questions. Read carefully and use mistakes to identify concepts to revisit; check the current dashboard for pass thresholds and retry rules.

Lastly, we have the programming exercises. They are presented with a notebook structure where the materials covered in the week are divided and explained, there are blocks of code for us to run without editing, and others in which we have to program a certain task.

They are a great material to go over the theory of the concepts again, and to learn how to implement them in practice. Their difficulty varies widely: in the first two courses the notebooks are quite easy, and most times the code is almost given to you entirely. Courses IV and V have harder notebooks with Tensorflow implementations of the networks that will challenge you.

Again, these notebooks are graded and you need a minimum grade to pass.

In the first courses, each week finishes with an interview called ‘Heroes of Deep Learning’ where Andrew speaks to a certain top personality of the field. This is a great addition and its definitely worth watching these videos. 

Certificates depend on meeting the provider’s completion requirements and having eligible access. They document course completion; an independent project helps demonstrate that you can apply what you learned.

If you want to get even more familiar with Artificial Neural Networks, check out these books:

• Deep Learning by Goodfellow et al: The book by the masters of Deep Learning, very theoretical and in depth text covering everything you need to know about ANNs.
• Grokking Deep Learning by Andrew Trask: A great introductory book that covers everything you need to know about Deep learning in a very illustrative and easy way, with lots of practical examples and diagrams. We love it for people that are looking to start in the world of Artificial Neural Networks.

Overview and Opinion

I think the format, with short videos, quick true or false test and notebooks is great. Longer videos tend to feel tedious and never ending, and in this manner you can really feel like you are making progress. This is also good for clearly separating the different contents, and allowing the students to go into more depth into any of them if they want to. 

Also, during the course, many practical cases are covered, and the lectures usually back up the theory with a real world example, which in my opinion helps a lot and is quite useful for consolidating the obtained knowledge. The third course is specially good in this sense, with real use-cases as exercises. 

I also like that despite having a quite solid mathematical foundations, the lectures insist more on the different aspects each optimisation /structure/ method solves, rather than on these maths. It also makes a clear differentiation between training and application, which is a fundamental separation to create in our way of thinking and become awesome Machine Learning practitioners. 

Overall, I think the specialisation is a great resource to learn about Deep Learning and artificial neural networks. It contains a lot of material, and the quizzes and notebooks really test your understanding of the contents. 

For me, it was definitely worth the time and money, and I hope I can put what I have learned to trial soon! If you are interested in doing it too, you can find it here. Enjoy it! 

You can find the link to the Coursera deep learning specialization here.

Note: If you think this course is too much for you to handle, do not worry, we have a section on Machine Learning books and another one on Machine Learning courses that will get you more than ready for tackling this spesialisation. Check them out!

Tags: Coursera Deep learning Specialization, Coursera Andrew Ng, Coursera Deep Learning, Deep Learning Specialization, Artificial Neural Networks, Coursera Machine Learning, Andrew Ng Machine Learning, Andrew Ng Coursera.