In the context of this lesson, is machine learning the same as artificial intelligence?
Answer
No, although similar in some ways, machine learning and artificial intelligence are different things. Machine learning is actually a subset of artificial intelligence, which is a broader field encompassing other concepts which include machine learning, robotics, and natural language processing.
Artificial intelligence can be described as when machines carry out tasks in an intelligent or smart way, based on set rules to solve certain problems. Artificial intelligence, or AI, makes decisions, learns, and solves problems similar to how humans would.
Machine learning, on the other hand, is a subset of artificial intelligence. It is when we give machines data and have them learn from that data on their own, without being explicitly programmed. Machine learning models learn from the data and try to make improvements to its predictions over time.
Machine learning is a technique that is used to accomplish artificial intelligence, but even without machine learning, artificial intelligence can still work. In a way, machine learning provides more efficient ways to make predictions on certain data, but is just a small part of artificial intelligence as a whole.
So, would it be correct to say that machine learning teaches a computer to interpret or analyze a dataset which allows the algorithm to essentially teach itself how to complete its task more efficiently according to the number of permutations run?
How would this ability translate to allowing a character in a virtual environment to learn from the users input without the need for additional coding? Would a neural network be the solution? And if so, would it apply in this example?
I would argue that it’s the other way around, that AI is part of ML.
AI uses ML to understand it’s environment/data intake.
Machine learning is quite a lot like how we learn, especially with the neural networks.
The more data we have, the more efficient our conclusions will be.
For instance, if we have the fact that 4 / 2 is not a decimal, we can’t infer much. However, if we receive lots of data, like all of the numbers between 0 and 100 (inclusive) and their quota from dividing by two, we can then infer that “even numbers are divisible by two in which they give us a whole number”.
Looking back, I think I meant that ML is a big subject on its own. Yes, I can see how it is a subset of AI, but it in itself is almost like a different topic. It is so big, in fact, that there is an entire branch dedicated to studying it. I don’t hear often about “AI” classes, but rather “Machine learning” classes, although I bet there are ones out there. Another big thing is that AI is really broad. Wikipedia goes on to say that
Leading AI textbooks define the field as the study of “intelligent agents”: any system that perceives its environment and takes actions that maximize its chance of achieving its goals.
So, theoretically we could program an AI in a game, that gets the value of an object (say, the player), and moves toward it.
It perceives its environment, and it takes an action that maximizes its chance of achieving its goal of reaching the player.
So really, I should have been saying that ML in itself is quite big, but yes, it is a subset of AI, “technically”.
