Blog: Artificial Intelligence 101
Artificial Intelligence is a term you have probably heard a lot nowadays. It seems that every new software product and every disruptive company uses AI as a secret solution that is just right up their sleeve. This has pushed this computer science discipline to have a massive growth in the world of technology this last few years.
But, what makes the AI so attractive to people today? Well, that’s because Artificial Intelligence is currently positioned as one of the best areas of computer science used to understand all the information that humans generate every day and that for humans is difficult to understand, and the thing about don’t understanding something is that it may sparks curiosity as well.
This post is focused on giving an introduction to this discipline, so if you’re interested in learning about this area and, most importantly, discovering how you can use it to improve the understanding of your data, you’re in the right place.
The learning in machines is only a simulation
When we talk about human learning we can distinguish between memorizing and real learning. We can say, for example, that memorizing a phone number is some “kind of learning”, but when we talk about learning, regularly we are talking about something else, something way more complex.
For example, when a little boy plays with other children, he can see how his actions have consequences and can impact other people and based on that all his social interaction will be influenced by that experience. Although he will not remember what exactly happened and he didn’t necessarily memorize that experience, but still he can recognize certain patterns in his environment and that patterns tell him how to react in future experiences. His knowledge goes further than just collecting and memorizing information, he is developing something we call intuition.
Based on that idea if we want to be more strict about defining what learning is, we will notice that actually machines don’t really “learn”, at least not as we humans do (at the moment). Despite this, Artificial Intelligence can simulate how humans learn by finding several relationships between different data types but through mathematical operations instead of empirical experiences.
A machine can learn in many different ways, depending on the algorithm that is being used, but in summary AI can learn data patterns by finding every possible combination in data and relate every combination with a possible result, and after all that, the AI will be able to use those relations to know how to react in future operations.
We live in a time in which the potential of AI has not been exploited just yet
Although artificial intelligence has been used to a greater extent in the last few years, we haven’t reached its full potential. Currently we know of three levels of Artificial Intelligence: ANI (Artificial Narrow Intelligence), AGI (Artificial General Intelligence) and ASI (Artificial Super Intelligence). Just now we are in the first level (ANI), an according to experts we will still need approximately between 5 to 30 years to reach the second level (AGI).
The Artificial Narrow Intelligence, as its name says, is the level in which AI is isolated to a narrow range of actions or capacities and can only mimic or simulate human intelligence. At this level we can find every intelligent assistant like Siri, Cortana and Google Assistant; in which we can interact with them but their main focus in centralized just in accomplishing certain tasks and they can’t do other activities with higher levels of difficulty like identifying criminal activity in your bank account. Also in this level, we can find every machine/deep learning algorithms developed in order to tackle extremely difficult problems such as identifying cancer through image analysis, but the algorithm will focus only on that task and it can’t do other activities as text analysis because it isn’t its goal, it’s not the purpose to which it was created.
It is important to say that at this level, ANI can’t surpass humans, because the machine still needs us to develop its “intelligence” but we can’t say the same for the next level. The Artificial General Intelligence is the stage in which machines have the same capacity as human intelligence, ergo they will be able to reason, plan to learn from experience and even handle complex concepts. At this level machine are capable of not only focusing on one single task but they can do it even better than humans.
And finally, Artificial Super Intelligence is the level in which machines will be smarter than all humanity combined. At this moment we don’t know much about this level, and maybe it can be difficult to imagine, but experts say that whenever we reach this level all technology will grow up in a way that we have never seen before and the vast majority of our problems will be solved.
Scope of Machine Learning in our time
As mentioned above, the use of AI has grown lately and it has reached a point in which we can find the use of AI almost everywhere.
Just to mention some of the most common uses for AI nowadays:
Natural language processing and speech recognition
This is one area of AI that we see most commonly used on the main intelligent assistants like Siri, Cortana and Google Assistant. With these tools is possible to make a petition and the assistant can “understand” the things we said and after that, answer based on the given petition. This can range from asking what time it is or look for information online, to schedule a flight on an airline.
Inside this area we can find a lot of uses, for example, facial and fingerprint recognition, used in several environments such as cell phones, monitoring systems through cameras, among other things. But also it can be used in medical environment for cancer recognition processes performed through magnetic resonance.
Recognition of criminal or suspicious activity
One of the most common uses of AI is in spam filters, this is so common that in some cases we don’t even notice that we are using AI technology in that moment. Also, another activity that uses the recognition of suspicious activity that has a bigger impact on our lives is in credit card fraud detection.
Can you imagine entering to Spotify and the first recommendation in your Daily Discovery is 50’s music? Either you love that music or you lived in that time, but if neither of those is the case, then it wouldn’t be nice that Spotify is recommending you that right?
AI technology is used repeatedly in order to know the tastes of the people and based on that provide to users the things they really need. For example provide you with the perfect mix made just for you, or tell you the best route to get to your favorite restaurant.
I want to learn AI
There are several ways to dive into this area, in our upcoming story we are going to provide you with some interesting resources that will help you to do that. Meanwhile, we want to give you a usefully post, “Stop Thinking. Start Learning” by Akhil Gupta, that focuses precisely on this topic and presents you with the most practical way to start learning this discipline.
If you’re interested in reading more about Artificial Intelligence, web development, among other things. I’ll be writing more articles in the near future.
I hope you enjoyed this article. I would love to know if you are using AI in your company or if you have any opinions about this content so please share your thought in the comments section below.
- Bell, J. (2015). Machine learning: hands-on for developers and technical professionals. Indianapolis: John Wiley & Sons.
- Brink, H., Richards, J. W., & Fetherolf, M. (2017). Real-world machine learning. Shelter Island: Manning.
- Every Day Big Data Statistics — 2.5 Quintillion Bytes of Data Created Daily. http://www.vcloudnews.com/every-day-big-data-statistics-2-5-quintillion-bytes-of-data-created-daily/
- Geitgey, A. (2016, December 23). Machine Learning is Fun! https://medium.com/@ageitgey/machine-learning-is-fun-80ea3ec3c471#.qrbj1qx0d
- Gupta, A. (2017, January 11). Stop Thinking. Start Learning. https://medium.com/data-science-group-iitr/stop-thinking-start-learning-cb74629bca3a#.ll9lgll5o
- Sysiak Pawel. (2017, July 27). AI Revolution 101. https://medium.com/ai-revolution/ai-revolution-101-8dce1d9cb62d
- Marsland, S. (2008). Introduction to machine learning: an algorithmic perspective. Boca Raton: Chapman & Hall/CRC.
- Müller, A. C., & Guido, S. (2016). Introduction to machine learning with Python: a guide for data scientists. Beijing: O’Reilly Media, Inc.
- Segaran, T. (2007). Programming collective intelligence: building smart web 2.0 applications. Beijing: O’Reilly.