Artificial Intelligence (AI) is now a common buzzword. You have heard it been used prominently in blockbuster movies like The Terminator, The Matrix etc. or due to the feud on AI-fuelled apocalypse between Elon Musk and Mark Zuckerberg. But if we leave aside the hyperbole, AI has made astronomical advances- cars can now be driven without drivers, games like Poker and Go have been mastered by supercomputers and most of our customer interaction and experience is being actually executed by AI. But still most of us are not clear when it comes to knowing the difference between Machine Learning (ML), Artificial Intelligence (AI) and Deep Learning (DL); we end up using the terms interchangeably or haphazardly. We will be sharing the key differences between them so that you can get a deeper insight into the world of AI and ML.
Artificial Intelligence (AI) The term “Artificial Intelligence” was first coined by the cognitive scientist John McCarthy in 1956. He described the technology as "Every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it."
In a broader term, AI portrays different ways a machine interacts with the world around it. Like a computer program playing a game of chess, to voice recognition systems like Amazon’s Alexa. AI is broadly divided under two categories: General and Narrow. General AI is the one which is at par with human- level, has pretty much all the characteristics of human intelligence and can perform a range of different tasks, whereas Narrow AI has some characteristics of human intelligence but is only skilled at one specific task. For example IBM's Deep Blue, which beat chess grandmaster Garry Kasparov at the game in 1996.
Machine Learning Not long after AI, Arthur Samuel coined the phrase Machine Learning in 1959 defining it as “the ability to learn without being explicitly programmed.” Basically, Machine Learning is a subset of Artificial Intelligence. ML focuses on learning rather than just computer programming; it is able to make predictions/adjustments based on the patterns identified by it. Its prime example is DeepMind’s Go-playing AI that defeated Ke Jie, world’s number one Go player, by learning and training itself from large data sets of expert moves.
Deep Learning Deep learning is the subset of Machine Learning and it takes computer intelligence to another level. It is an expensive proposition, as it needs large amounts of data to train itself. Basically, it imitates human brain’s connectivity in classifying the data set and from there it draws the correlations between them. Since there are a lot of underlying parameters involved, there are greater chances that the results may not be accurate in the beginning. For example, if a deep learning algorithm is instructed to “learn” how dogs look like, it will take a massive amount of picture data in order for the machine to learn minor details about the dogs and distinguish them from other animals. Facebook’s photo recognition software is now able to tag your friends with 98% accuracy, all because of deep learning.
Future: All of these three technologies are going to play a crucial role in the development of technology (in a narrow) and humanity (in the larger sense). Many commercial or socio-economic problems that seem impossible to solve today would be near a feasible solution in the future because of this trifecta of technologies. Stay tuned.
This article is sponsored by Lampix. Visit us at www.lampix.co today.