Updated: Apr 21, 2020
Two-letter abbreviations seem to be everywhere these days, such as ML, AI, DL (deep learning), VR, and XR. These might be some of the hottest buzzwords (buzz-abbreviations?) of the decade but the truth is that most people don’t actually know the difference between many of these, especially AI, Ml, and DL.
Why is this important?
This is very important in research as mixing these crucial terms up throws off the reader in the concept being presented in the research. In other words, the research starts to change meaning with the intertwining of these buzzwords as the models that originate from, for example, deep learning are different than just general AI.
To get down to the core of this question, we must dive deep into the definitions.
What is artificial intelligence or AI?
Artificial intelligence is simply the act of adding human intelligence to machines. Before AI revolutionized technology, every piece of technology ran on an algorithmic basis. In other words, machines were not adaptable to certain situations, rather ran strictly on instructions. Throughout history, this notion created a social, perceptive superiority over machines. However, scientists around the globe wondered how much better technology would be if it thought like us, sparing the AI revolution.
AI powered machines are usually categorized in 2 ways - general and narrow. An example of a general AI machine is a robot or software that can detect your hand movement. An example of narrow AI is something that classifies images on Pinterest.
The difference for artificial intelligence is that it encompasses both deep learning and machine learning as machine learning and deep learning build on and enhance AI powered machines, but aren’t AI themselves.
What is machine learning?
The name is self explanatory. Machine learning is the enhancing a computer or machine’s ability to learn. Dr. Michael J. Garbade of Towards Data Science defines the purpose of ML as something that, “enables machines to learn by themselves using the provided data and make accurate predictions.”
In machine learning, input data is passed to a training algorithm. This central training algorithm essentially defines machine learning. Experts develop variations of training algorithms to allow machines to learn better given the current data. With the knowledge the machine gains from the input data, it outputs predictions and patterns observed based on the input data and the interpreting machine learning algorithm.
The difference between machine learning and artificial intelligence is that, as Bernard Marr of Forbes describes, machine learning is specific to inputting data into a machine so it can learn by itself, while artificial intelligence is a broad term that’s main goal is to make a machine smarter.
What is deep learning?
Last, but definitely not least, is deep learning.
As Jason Brownlee of Machine Learning Mastery puts it, “Deep learning is concerned with algorithms inspired by the structure and function of the brain called artificial neural networks.” The entire concept behind machine learning, artificial intelligence, and deep learning is emulating human intelligence into machines. Therefore, deep learning uses a big concept that makes our intelligence, the millions of networks in your brain that are bursting neurons every second.