I am interested to learn about machine learning, which will help me in my research. I have no experience in this area and looking for a good read for beginners. Thanks in advance.
(Machine@ Learning)Machine Learning by Andrew Ng on Courserahttps://www.coursera.org/learn/machine-learning
How does a total beginner start to learn machine learning if they have some knowledge of programming languages?https://www.quora.com/How-does-a-total-beginner-start-to-learn-machine-learning-if-they-have-some-knowledge-of-programming-languages/answer/Karlijn-Willems-1ML
Andrew Ng's course on Coursera is an excellent place to start. For more in-depth knowledge please refer to the books Introduction to Statistical Learning and Elements of Statistical Learning.
(Machine@ Learning)Machine Learning by Andrew Ng on Courserahttps://www.coursera.org/learn/machine-learning
How does a total beginner start to learn machine learning if they have some knowledge of programming languages?https://www.quora.com/How-does-a-total-beginner-start-to-learn-machine-learning-if-they-have-some-knowledge-of-programming-languages/answer/Karlijn-Willems-1ML
Sanwar Ahmad, can you tell us a bit about the task you intend to perform using ML, and if you consider using Soft Computing or Hard Computing (classic) algorithms?
If you want to use Soft Computing algorithms in your research, more specifically fuzzy logic (your projects show that you have experience working with fuzzy logic and fuzzy sets), none of the suggestions given to you will help you because they focus on Hard Computing and exclude Fuzzy Logic based algorithms. If you want to focus on Hard Computing algorithms for ML, then all the previous suggestions are good.
A basic reference for fuzzy sets and fuzzy logic is:
Fuzzy Sets and Fuzzy Logic: Theory and Applications 1st Edition, George J. Klir & Bo Yuan - http://www.amazon.com/Fuzzy-Sets-Logic-Theory-Applications/dp/0131011715
As you have previously applied Fuzzy Logic in your research works, I suggest you study ML using classic/Hard Computing techniques (all of the references previously given are good) and then look for proposals in the literature focusing on the same task but based on fuzzy logic.
For example, one of the most commonly performed tasks in ML is classification (consider what is done by Facebook every time you post a new photo: an algorithm is used to search for areas in the new picture that might contain the face of a person and suggests you mark that area as an existing user/profile of the social network). These tasks can be done using the classic K-nearest neighbour (KNN) algorithm or the Fuzzy version of KNN. The same is valid for many other techniques and tasks in ML.
Summing up, although the task might is the same, the paradigms are quite distinct.
Thanks Marcos Evandro Cintra for the details. I do had previous research experience in Fuzzy logic. Currently I am working on inverse problems in tomography problems. I am developing a hybrid approach for image reconstruction, combining the deterministic and statistical approach. I got interested in ML as it is also used to reconstruct images. Thanks again for sharing the application of ML using Fuzzy logic.
I would recommend two books: Pattern recognition by Christopher Bishop and Deep Learning by Ian Goodfellow et al. These two books unravel and expatiate on machine learning with lucid illustrations.