Using the most adopted classification, ML is part of AI, addressing one of the cognitive areas into the Artificial Intelligence wide scope: learning.
Into ML, we can found different learning models with corresponding associated algorithms. Artificial Neural Networks (ANN aka NN) are the basis of Deep Learning models. Therefore, all learning models using Artificial Neural Networks can be grouped as Deep Learning models. Deep Learning is part of Machine Learning, not something different or classified in the same level.
Neural network is a machine learning method like other ML methods.
We have several ML algorithms and each of them has its own logic.
For example, the support vector machine (SVM) tries to maximize a margin between classes in training data. KNN just uses nearest neighbors in training data for labeling new samples, etc.
There are 3 subsets in ML: Supervised, Unsupervised and Reinforcement Learning. For now, we use Neural Networks for both supervised and unsupervised learning.
Machine learning also contains some other techniques such as ensemble learning, transfer learning, etc.
Using the most adopted classification, ML is part of AI, addressing one of the cognitive areas into the Artificial Intelligence wide scope: learning.
Into ML, we can found different learning models with corresponding associated algorithms. Artificial Neural Networks (ANN aka NN) are the basis of Deep Learning models. Therefore, all learning models using Artificial Neural Networks can be grouped as Deep Learning models. Deep Learning is part of Machine Learning, not something different or classified in the same level.
Artificial intelligence includes both machine learning and deep learning. as more precisely, deep learning is a part of machine learning. as if you see the hierarchy of the subject artificial intelligence, you will find all related to one another like computer architecture which is defined as kernel, shell and etc.
Indeed machine learning includes both supervised and unsupervised learning.
we have to train and test the system for the classification.
In supervised learning, we can use various types of algorithm with feed forward neural network:
1)backprogation
2)svm
3) k-nn
4) linear regression
similarly, we can use unsupervised which are:
1) k-means
2)principle component analysis
where as deep learning is used specifically in image recognition, speech recognition, and etc with the help of following :
1) deep believe network
2) deep neural network
3) recurrent neural network
Deep learning is used mainly for adaption of self learning capability of the system.