How is Machine learning different from Artificial Intelligence? What are the prerequisites for learning machine learning?

Answer 1

Think of Artificial Intelligence as being the “goal” and Machine Learning as a a collection of subsets or a means to get to that “goal”.

Consider this; you’re walking in the woods and you collect some flowers. You collect some data on the flowers and make notes on how many petals each has. You also note the length and width of the petals. Perhaps also noting the color. Now, with this data you author some code around KNN (K Nearest Neighbor) and use the data you collected to help classify future flower findings. This is ML.
In an effort to speed up the collection of data you write up some convolutional neural network code that looks at images and tries to figure out what is a flower, how many petals does it have, how big are the petals and what color are they. After providing your code with a few thousand images of flowers, it achieves a 99% success rate of accuracy on the data you require. Now you have a means to find things that look like flowers and a way to classify them. This too is ML.

Continue writing ML codes to further the collection of data on flowers. At some point you might get an advanced expert system. Maybe the code will stop ignoring trees and devise its own question: Why am I only classifying flowers when I could be looking at these giant trees? That would be a high level of AI.

My last example might sound a bit like science fiction NOW, however I’m quite sure our grandparents would have never conceived of the possibilities of the smallest and simplest machine language code and the potential it would lead to.
In summary, ML would encompass many various methods to classify data and make predictions. AI could be the result of many ML codes which would be capable of asking/answering its own questions with minimal to no supervision.
🇺🇸