Some time ago, while browsing through the latest AI news, I found a company that claims “machine learning and advanced artistry to collect and analyze hundreds of data touch points to improve the user experience in a mobile app.” intelligence”has been used. On the same day, I read about another company that predicted customer behavior using “a combination of machine learning and AI” and “AI-powered predictive analytics”.
(I will not name companies to avoid shaming them, because I believe their products solve real problems, even if they are marketing it in a deceptive way.)
There is a lot of confusion about artificial intelligence and machine learning. Some refer to and use AI and machine learning as synonyms, while others use them as separate, parallel technologies. In many cases, people who speak and write about technology do not know the difference between AI and ML. In others, they deliberately ignore those differences in order to generate publicity and enthusiasm for marketing and sales purposes.
Like the rest of this series, in this post, I will try to explain the differences between Artificial Intelligence and Machine Learning to help you distinguish fact from fiction with respect to AI.
We will start with machine learning, which is the easy part of the AI vs. ML equation. Machine learning is a subset of Artificial Intelligence, one of many ways of doing AI. Machine learning relies on defining behavioral rules by examining and comparing large data sets to find common patterns. This is an approach that is particularly efficient for solving classification problems.
For example, if you provide a machine learning program with a lot of X-ray images and their associated characteristics, it will be able to aid (or possibly automate) the analysis of X-ray images in the future. The machine learning application will compare all those different images and find out what are the common patterns found in images labeled with similar traits. And when you provide it with new images it will compare its contents with the pattern it has assembled and will show you how the images may have had any symptoms that he has studied before.
This type of machine learning is called “supervised learning”, where an algorithm imparts training on human-labeled data. Another type of ML unheard learning relies on giving the unlabeled data to the algorithm and letting it find patterns on its own. For example, you provide an ML algorithm with a continuous stream of network traffic and let it learn by itself what basic, normal network activity is, and what external and potentially malicious behaviors are occurring on the network.
Reinforcement learning, the third popular type of machine learning algorithm, relies on providing an ML algorithm with a set of rules and constraints and letting it learn how to best achieve its goals. Reinforcement learning usually involves a type of reward, such as gaining points in a game or reducing power consumption in a facility. The ML algorithm tries its best to maximize its rewards within the provided constraints. Reinforcement learning is well-known in teaching AI algorithms to play various games like Go, Poker, Starcraft and Dota.
Machine learning is attractive, especially as it is more advanced subsets such as deep learning and neural networks. But this is not magic, even if we sometimes have problems understanding its internal functioning. At its heart, ML is the study of data to classify information or predict future trends. In fact, while many people like to compare deep learning and neural networks to the way the human brain works, there is a huge difference between the two.
Bottom Line: We know what machine learning is. It is a subset of Artificial Intelligence. We also know what it can and cannot do.
We don’t exactly know what AI is
On the other hand, the scope of the term “artificial intelligence” is very wide. According to Andrew Moore, dean of computer science at Carnegie Mellon University, “artificial intelligence is the science and engineering of making computers that behave in the way we thought human intelligence needed.”
This is one of the best ways to define AI in a sentence, but it still shows how broad and unclear the field is. For example, “until recently” is something that changes over time. Many decades ago, pocket calculators were considered AI, because computation was something that only the human brain could do. Today, the calculator is one of the best applications that you will find on every computer.As Zachary Lipton, editor of Almost Correct, explains, the term AI is “aspirational, a dynamic goal based on capabilities that humans possess but which machines do not.
AI also includes a lot of technologies that we know about. Machine learning is just one of them. Other methods of AI were used in earlier works, such as good old-fashioned AI (GOFAI), which is an if-then rule that we use in other applications. Other methods include A *, fuzzy logic, expert systems, and more. In 1997, AI Deep Blue, who defeated the world’s chess champion, used a method called tree search algorithm to evaluate millions of moves at every turn.
A lot of the references made to AI relate to general AI, or human-level intelligence. You see that kind of technology in sci-fi movies like The Matrix or 2001: A Space Odyssey. But we still don’t know how to create artificial intelligence that equals the human mind, and intensive learning, the most advanced type of AI, can rival the mind of a human child, let alone an adult. This is perfect for narrow tasks, not general, for abstract decisions, which is not a bad thing.
AI as we know it today is the epitome of Siri and Alexa, the weirdly accurate film recommendation system that powers Netflix and YouTube, which are used by algorithms to create hedge fund micro-trades that make millions of dollars every year Rakes in Huh. These technologies are becoming increasingly important in our daily lives. In fact, they are augmented intelligence technologies that enhance our capabilities and make us more productive.
Bottom Line: Unlike machine learning, AI is a dynamic target, and its definition changes as its related technologies become more advanced. The definition of what AI is and cannot be easily opposed, while the definition of machine learning is very clear. Maybe in a few decades, today’s state-of-the-art AI technologies will be considered dumb and monotonous as calculators are for us right now.
So if we go back to the examples mentioned at the beginning of the article, what exactly does “machine learning and advanced AI” mean? After all, aren’t machine learning and deep learning the most advanced AI technologies currently available? And what does “AI-powered predictive analytics” mean? Doesn’t predictive analytics use machine learning, which is a branch of AI anyway?
Why do tech companies like to use AI and ML interchangeably?
Ever since the term “artificial intelligence” was coined, the industry has gone through many ups and downs. In the early decades, there was a lot of hype around the industry, and many scientists promised that human-scale AI was around the corner. But unfulfilled promises led to a general disenchantment with the industry and led to an AI winter, a period where funding and interest in the area waned significantly.
Later, companies tried to differentiate themselves from the term AI, which was synonymous with unfounded propaganda, and used other terms to refer to their work. For example, IBM described Deep Blue as a supercomputer and explicitly stated that it did not use artificial intelligence, while technically it did.
During this period, other terms such as Big Data, Predictive Analytics and Machine Learning began to gain traction and popularity. In 2012, machine learning, deep learning and neural networks made considerable progress and began to be used in more and more fields. Companies suddenly started using the terms “machine learning” and “deep learning” to market their products.
Deep learning started doing tasks that were impossible to do with rules-based programming. Areas such as speech and facial recognition, image classification and natural language processing, which were in very crude stages, suddenly took a huge leap forward.
And maybe that’s why we are seeing a change back in AI. For those who were used to the limitations of old-fashioned software, the effects of deep learning seemed almost like magic, especially since neural networks and deep learning are entering some areas, off to computers. -Limit was considered. Machine learning and Deep learning engineers are earning 7-digit salaries, even if they are working in nonprofits, which shows how hot the sector is.
Add to this the misleading description of the neural network, which claims that the structure mimics the work of the human brain, and you suddenly realize that we are again moving towards artificial general intelligence. Many scientists (Nick Bostrom, Elon Musk…) started warning against an apocalypse in the near future, where super-intelligent computers lead humans to slavery and extinction. The fear of technical unemployment arose again.
All these elements have helped rekindle the enthusiasm and publicity surrounding artificial intelligence. Therefore, sales departments find it more profitable to use the ambiguous term AI, which carries a lot of baggage and experiences a mystic aura rather than being more specific about the kind of technologies they employ. This helps them to oversell or remarket the capabilities of their products without being clear about their limitations.
Meanwhile, the “advanced artificial intelligence” that these companies claim to use is usually a type of Machine learning or some other known technology.
Unfortunately, this is something that technical publications often report without intense scrutiny, and they are often accompanied by AI articles with images of crystal balls and other magical representations. This will help those companies to publicize about their offerings. But down the road, as they fail to meet expectations, they are forced to hire humans to meet the shortcomings of their AI. Finally, they can create mistrust in the region and start another AI winter for short term gains.
0 Comments