0
$\begingroup$

The answers to this Quora question say it's OK to ignore machine learning and start right away with deep learning.

Is machine learning required or is useful for understanding (theoretically and practically) deep learning? Can I start right away with deep learning or should I cover machine learning first? In what way machine learning useful for deep learning? (leave the mathematics part - I'm ok with it).

$\endgroup$
  • 1
    $\begingroup$ By rephrasing the Quora question you essentially changed its meaning. The original question asks on whether or not to invest several years in learning traditional ML techniques. Deep Learning is a subfield of Machine Learning. Even if you picked up a DL book (e.g. goodfellow's) you'd see a lot of ML background. $\endgroup$ – Djib2011 Nov 22 at 9:14
1
$\begingroup$

enter image description here

  1. Is Machine Learning required or is useful for understanding (theoretically and practically) Deep Learning?

NO

Deep learning is itself a huge subject area with serious applications in NLP, Computer Vision, Speech and Robotics. You should learn deep learning from scratch like understanding forward propagation, back propagation, how weights are updated etc.. instead of using high level frameworks like keras, pytorch. It's OK to use them once you understand the basics to save time and code complexity, but remember "surely" you don't need machine learning for that.

Since you are familiar with the mathematics part, I would suggest you to straight away jump into Deep Learning. Note that deep learning is inspired by how the brain works.

  1. Can I start right away with Deep learning or should I cover Machine learning first?

Yes you may start right away, start with the hello world problems "MNIST DIGIT Classification" if you know little image processing. Start with a simple neural network model from scratch, then use keras (very easy) and then proceed to CNN ... You may start with simple problems in other fields too (NLP, Speech) I suggest Andrew Yangs, course in Machine Learning (within this he explains a neural network model for MNIST I guess).

  1. In what way machine learning useful for Deep learning?

You will understand that in machine learning you sit down and find useful features in the dataset yourself, but in deep learning it happens automatically (Learn Deep learning in detail and come back and read this, you will understand exactly what I mean!) If you learn Machine learning and then go to deep learning, you will realise that it was unnecessary .if you interested in this field of AI, jump into deep learning right now!

$\endgroup$
4
$\begingroup$

Deep learning is part of machine learning.

  • You will miss out useful information if you ignore machine learning.
  • You are ok to start your work in machine learning with deep learning and neural networks. You have to start somewhere and starting with a strong and successful method is resaonable, especially if you need to be able to produce good results quickly.
  • You will learn essential machine learning stuff while reading about deep learning.
  • The deep learning tutorials and other learning materials you will be reading may not be telling you that what you are learning also applies to other machine learning methods but you will be learning lot's of stuff that applies more generally. You will be studying some machine learning whether you want to or not.
  • If you have plenty of time a more broad view will help understanding. Still, there is no need to wait with deep learning to after mastering some other methods.
  • Broader knowledge helps you to relate and memorise concepts and be more aware of potential issues, especially issues that are rarely discussed in the deep learning community. Such knowledge and experience will be most useful when trying to apply deep learning to new problems or if trying to make substantial changes.
$\endgroup$
0
$\begingroup$

I would argue definitely since it is a bit sequential. (i) Start off applying basic machine learning concepts such as regression, classification and generalization techniques etc. to real world problems. (ii) you will soon realize the limitations of those techniques.(iii) Take your learning to the next level by learning and applying deep learning concepts specially if the issues are around image classification or NLP. As mentioned by @Joachim Wagner you will not only miss out on useful info but there will a huge gap in our learning. Hence, i would suggest learning concurrently or ML first otherwise DL will become black box of black box.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.