We all have heard of AI, Machine Learning and Deep Learning in various articles blogs and books, but
What is artificial intelligence, machine learning, and deep learning? How do they relate to each other?
Artificial Intelligent started in 1950’s with the idea of making computer’s think.
AI can be defined as :
“The effort to automate intellectual tasks normally performed by humans”
AI is a general field comprises of Machine Learning and Deep Learning. It also also comprises of fields that doesn’t require any learning at all such as; Early Rule based systems or Expert systems.
During the early days of Artificial intelligence many of the expert believed human intelligence level can be achieved simple by handcrafted Explicit rules in rule based systems. This approach is known as symbolic AI. This approach was dominant paradigm during 1950’s to 1980’s during the rise of Expert systems.
Symbolic AI performs well on defined and Logical problems such as playing chess. It does, however, not perform well on complex and fuzzy problems such as; image classification, speech recognition and language translation.
Machine Learning arises from this very question:
“Could computers go beyond and actually learn how to perform a specified task. without handcrafting explicit rules for the task. Can a computer automatically learn the rules by looking into the data ?”
Symbolic AI v/s Machine Learning
In Classical Programming, Symbolic AI User would input Data and rules which will process these Data to produce answers.
In Machine Learning, User would input Data and expected Answers and the output produced is Rules. These rules can then be applied to new data to produce answers.
Machine Learning systems are Trained rather than explicitly programmed. Machine Learning started flourishing in 1990’s. It has become the most popular sub field of AI, Due to the availability of faster hardware and larger datasets.
Deep Learning is specific sub field of Machine Learning, a new way of learning from data which puts an emphasis on learning successive “layers” of increasingly meaningful representations.
In deep learning, these layered representations are learned via models called “neural networks”.
The “deep” in “deep learning” is not a reference to any kind of “deeper” understanding achieved by the approach, rather, it simply stands for this idea of successive layers of representations — how many layers contribute to a model of the data is called the “depth” of the model.
Modern deep learning often involves tens or even hundreds of successive layers of representation — and they are all learned automatically from exposure to training data.
Read also: Why should you learn Machine Learning?