Chevron Left
Back to Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization

Learner Reviews & Feedback for Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization by DeepLearning.AI

4.9
stars
62,820 ratings

About the Course

In the second course of the Deep Learning Specialization, you will open the deep learning black box to understand the processes that drive performance and generate good results systematically. By the end, you will learn the best practices to train and develop test sets and analyze bias/variance for building deep learning applications; be able to use standard neural network techniques such as initialization, L2 and dropout regularization, hyperparameter tuning, batch normalization, and gradient checking; implement and apply a variety of optimization algorithms, such as mini-batch gradient descent, Momentum, RMSprop and Adam, and check for their convergence; and implement a neural network in TensorFlow. The Deep Learning Specialization is our foundational program that will help you understand the capabilities, challenges, and consequences of deep learning and prepare you to participate in the development of leading-edge AI technology. It provides a pathway for you to gain the knowledge and skills to apply machine learning to your work, level up your technical career, and take the definitive step in the world of AI....

Top reviews

AS

Apr 18, 2020

Very good course to give you deep insight about how to enhance your algorithm and neural network and improve its accuracy. Also teaches you Tensorflow. Highly recommend especially after the 1st course

AB

Aug 26, 2021

Amazing course which focus on the theoretical part of parameters tuning, but it needs more explanation of Tensorflow, as I felt a little lost in the last project. Except that, it is an amazing course.

Filter by:

6126 - 6150 of 7,216 Reviews for Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization

By Juan P A A

•

Jan 27, 2020

The contents are actually good, and it doesn't require a very extensive prior knowledge, so it's even suitable for people with little experience in programming or math. However, despite being a course that has been out for over 2 years, there are still some subtitle issues (in English), and typos on a clarification slide.

By John H

•

Aug 24, 2017

Well explained..sometimes jumps a bit. I felt lost a couple of times. But I got through it and I'd say this is deifnitely one of the top courses out there.

If they included some optional videos on how this could relate to having a career in this area that'd be very helpful (i.e. what level we need to be able to code at).

By Anmol K

•

Jun 16, 2020

This course continues to build on foundations from course 1 of the specialization. Hyperparameter tuning and Regularization methods are quite imperative for optimizing ML models. This course covers these concepts in addition to providing a good foundation for Tensorflow library. Overall, a good course by Prof. Andrew!

By Katsiaryna R

•

Jul 24, 2019

The course was very helpful as now I understand optimization techniques and all the parameters of neural networks. Unfortunately, the course has not answered my question how to tune the whole bunch of hyperparameters from the scratch, what is the correct order and logic of the full ANN tuning, not just one parameter.

By Srinivas R

•

Sep 22, 2017

A very good follow on to the first course with continued excellent organization and hands on assignments that give you practical exposure to working on deep learning problems including a basic introduction to Tensorflow along with practical guidelines on Hyperparameter tuning among other deep learning related topics.

By uday r

•

Mar 11, 2018

Hi,

This course does a really good job in introducing the optimization techniques. Prof. Andrew Ng has structured his lectures well.

Can I kindly suggest that this course can incorporate, for each optimization, 1 scenario that is applicable & 1 that isn't? That will emphasize the scope of the optimization.

Thanks,

Uday

By Shawn W

•

Oct 2, 2017

Still some errors in the assignments, particularly in Week 3. Otherwise, a good course. A lot of good topics for practical use when building real-life neural networks. Some seem fairly cutting-edge. Good, brief, introduction to TensorFlow at the tail-end of Week 3 and in the programming assignment for that week.

By Raghav B

•

Dec 26, 2018

The course content is really great and the theoretical concepts (or their intuition to a larger extent) have been explained pretty well. But there are some errors in the programming exercise on Tensor Flow which makes it confusing since the people who take this course are new to both deep learning and tensor flow.

By Chris A

•

Apr 29, 2018

This is a great course. The only reason for not giving it 5 stars is the notebook platform for grading coding assignments. It is flawed in that attempts to save often error out so a submission often doesn't contain the latest state of the code. This causes sections to be graded incorrectly - very frustrating.

By Dmitrijs T

•

Jan 10, 2020

Course material and Lectures is perfect!

May be would like to have less supervised programming assignments with less hints of how to implement code as it was too easy! May be it would be good to have pretty guided assignments during main part of the course, but something more individual and demanding at the end.

By Carles S F

•

May 28, 2018

I think it is important to understand the basics and that is why it is really cool that they show you how to implement a neural net from sratch. Moreover, the last part on tensorflow shows you how to do it in real life. Nevetheless, a lot of work remains to be done to learn properly how to use tensorflow and NN

By John H

•

May 8, 2020

Andrew Ng is great at explaining the theory and practical aspects of DL concepts. I applaud him for making the lectures so accessible. I also really liked that he provided what is typically done by practitioners. My only feedback I have is that the quiz and programming assignment could have been more rigorous.

By Yunlin Z

•

Sep 7, 2017

the programming assignments are too easy. Though I understand that they're supposed to guide someone who may be a total beginning in ML and DL, I feel there is still too much hand-holding by marking exactly the changes need to be done and the formula either embedded in the comments or in the description above

By Bhavul G

•

Apr 3, 2018

Brilliant. The way this whole course was structured, the correctness of everything, and the amount of thought and preparation that has gone into this is amazing. Thank you.

One suggestion : for those who would want to fiddle with math, there could be optional links to read / understand derivations and stuff.

By Anirudh R

•

Jun 6, 2020

I understood in detail about the various hyper parameters and how to tune them in order to obtain the best results. I learnt about the problem of over fitting and how to solve it. I also got a glimpse of the tensor flow framework and understood how it makes the process of creating neural networks easier.

By Austin G

•

Dec 15, 2018

Great until the last weeks exercise. It definitely needs to be improved. Some of the instructions do not make sense and there are errors in the outputs that make it confusing to follow. However, I did complete it in 1 hour where it said it would take 3. That being said, all videos were amazing as always.

By JATIN S

•

Jul 27, 2020

The content for the first two weeks can be thoroughly practised using the programming assignments but for the 3rd week there is a huge gap in the practise material provided.The majority of the content taught during this week is not covered in the programming assignment.Hope this gets seen and rectified.

By Minseok L

•

Jan 31, 2018

While the first course in Deep Learning Specialization gives us the fundamental insight, this course shows us the practical aspects such as hyper-parameters tuning and mini-batching . I think it is essential for beginners to take this course if they actually want to apply Deep Learning to their domains.

By Xinyu L

•

Jul 9, 2020

Generally great course! But I think more tutorials about Tensorflow are needed. I suppose by using the current tutorial in this course, no one except those who have used Tensorflow before, is able to apply Tensorflow to their own work. The programming exercise is more like a

shallow experience so far.

By Ali K

•

Feb 27, 2020

The course was useful for me as a beginner in the deep-learning domain. The instructor was very clear and easy to understand. In order for the student to easily grasp some concepts, useful intuitions were provided. However, being a mathematics student I would have appreciated a more thorough analysis.

By Karthik V

•

Aug 31, 2018

Very good overview of optimizer functions and hyperparameter search. I felt the Tensorflow portion could be better with more emphasis put on what the TensorFlow programming construct is as thinking in terms of computational graph and functional programming aspect could be covered as a general overview

By Nimesh P

•

Sep 7, 2020

The courses are good and impart knowledge from the experience of Dr. Andrew Ng. However, I wish they had more of "After the lecture" 1-2 question quizzes [they have such quizzes designed in course 1 of deep learning specialization].

These quizzes bring up the level of course (and learning) infinitely.

By Ankur C

•

Jan 22, 2018

Again, A very good course , it focuses on building the concepts from fundamentals. I got very intuition of the way neural nets work.

The Jupyter notebooks are self explanatory and comprehensive. Though I think programming could have been a little tougher, could have used a fully hands on assignment.

By Johan W

•

Oct 30, 2017

Starting to get somewhere.. This course contains a bit more interesting topics and I did not feel it was as overlapping to the Ng Machine Lerning course as the first course in the Deep Learning specialization was. Also, finally, a programming task is done using a deep learning framework (TensorFlow)

By John S

•

Nov 11, 2018

Good material as always from Andrew Ng. There is a lot more "eyes-forward lectures" before the coursework on this one, which didn't suit my my personal learning preference as well (I prefer to get a lot of practise in when getting to grips with new concepts). Good course though, worth the effort.