Data Science and Machine Learning are emerging as one of the hottest topics in the advancing industrial or technology domain. Almost every company, ranging from the top tech giants like Google, Microsoft, or Apple to newly emerged startups like Uber, or Swiggy, has already started building the infrastructure for Machine Learning support. If we take the case of Uber, the business revolves around the usage of Machine learning. You can find the details here. The promise it has shown to the people is incomparable and, people are expecting more from it.
No doubt that it has shown tremendous potential till now, but it also has created hype or so-called rumors among the people. We sometimes overemphasize the capabilities of technologies, and Machine Learning models are not an exception here. In this article, we will discuss the top ten most common misconceptions which are so popular that every one of us must have come across at least once in our Machine Learning journey or broadly saying the computer science journey.
This is one of the most famous misconceptions that almost every early-stage learner must have heard. But in reality, Deep Learning is just a part of Machine Learning, and Machine Learning is just a part of Artificial Intelligence. This can also be represented in the below Venn diagram. The apparent difference with the example is covered here.
Machine Learning models have opened new verticals of technologies that automate the process and improve traditional works' quality. It is not a complete myth but a half-true fact. Every technology is growing, and we need to evolve our technical skills. There is one famous statement of Dr. APJ Abdul Kalam.
“It is not the unemployment which is a major problem; it is the question of ‘unemployability’ which is a bigger crisis”.
If we have the skills but are unemployable, it means that skill-set is outdated, and we need to build our technical expertise around the newly developed tools or technologies. If you see the data from different industries, Machine Learning engineers or Data Scientists have become the highest paying jobs globally. The average salary of Machine Learning experts and Data Scientists is more than 50K USD which is the highest in the Information Technology industry. Machine Learning has increased and added the number of employments rather than reducing it.
There is no doubt that Machine Learning models can predict the future, but this statement is again half-true. This always comes with an additional condition which we generally ignore. Machine Learning models can predict the future only if future events have similarities with past events. Or formally, we can say if history repeats itself.
For example, there are machine learning-based forecasting models, using which we can forecast predictions like stock price based on the previous day prices, weather conditions based on similarity with prior dates, production estimates based on the production of prior years. There are many more examples. But, if you ask them to predict the patterns what was not seen earlier while building the machine learning model, they will fail to predict future things.
In the present stage, Machine Learning models are NOT capable of solving all the issues of this world. One of the most famous quotes about machine learning says :
“A baby learns to crawl, walk and then run. We are in the crawling stage when it comes to applying machine learning.” ~Dave Waters
Algorithms developed till now are specifically designed to solve a particular type of problem statement. Companies like Facebook and Google have started visualizing Machine Learning from a “Non-Supervised Learning” perspective where there will be no labeling of data available, and machines will learn from themselves. They have shared strong evidence that Supervised Learning can never become the future because of the dependency on the annotation or labeling of a massive amount of dataset.
This is also one of the most common doubts in the field of Machine Learning and Deep Learning. People think that one machine learning algorithm is sufficient to solve all the problems that exist in this domain. There is the misconception that algorithms have evolved, and hence further evolvements in algorithms will have enough power to solve earlier algorithms' problems. But this doesn't seem right. The Logistic Regression algorithm is not able to solve the regression problems.
The fact that algorithms have evolved is entirely correct, but this evolution is made with a focus to solve wholly different or advanced problems. So the choice of algorithm is totally dependant on the problem statement that we want to solve.
This misconception is wildly used and spread via sci-fi movies. Machine Learning algorithms, which are developed till now, are not even able to make self-made decisions, and we think that robots will achieve human-level intelligence. In other words, we can definitely say that humans control it.
Many scientists are also working in Artificial General Intelligence (AGI) to make machines able to take decisions independently. When this technology is developed, we will say that humans will make super-intelligent robots shortly.
This is one of the most interesting misconceptions among readers. Earlier, we said that ML had not evolved that much, but for this case, Machine learning has evolved to that point where it does not require any kind of human interventions. Technologies have evolved to the point where Machine Learning or Deep Learning algorithms can automatically generate short news content or articles based on the data available over the internet, which requires zero intervention of humans.
So, with advanced algorithms, we can altogether remove the human involvement after training the particular machine learning model.
There are some automatic videos generated with the “deep fake” technology where fake faces of any human being can be used. They look so real that some people believe that those videos actually contain persons present in the video. So, human intervention may be reduced to zero in the case of machine learning.
In Deep-learning, we can say that the model's accuracy depends on the amount of data we feed to our model. But the ability of our machines, hardware, and computation power will be the biggest bottlenecks here. We will not be able to provide all the data to our machine learning model at once.
Sometimes, mining the features from the same dataset may not be helpful. Models can be overfitted or biased towards one or more features because the new extracted feature is highly correlated to the existing features.
This is also one of the most prevalent misconceptions that persist evn in many advanced learners. The role of Machine Learning algorithms is to provide you higher modularity rather than higher accuracy. There may be situations where any ML algorithm will not beat the existing traditional algorithms in terms of accuracy. Still, it can certainly beat the standard algorithm if it will perform better in unexpected scenarios or it has to consider the exceptional scenarios.
For example, suppose there is one traditional algorithm to calculate our target variable for our supervised learning model. Our machine learning model will try to reach 100% accuracy with respect to that conventional algorithm. As this will never be completely 100%, we can say our Machine Learning model will never beat the conventional approach. We must be thinking that then why ML? Because it may be possible that the overall complexity and computation power required in the conventional algorithm is higher than the machine learning models. In such a case, replacing traditional algorithm with machine learning models will be highly beneficial.
While struggling with many Machine Learning research works, we might have gone through several research papers that justify using traditional methods over machine learning algorithms by saying that Machine Learning technology is hard to implement. But the reason for that statement is that they are domain experts in the traditional techniques, and their knowledge in ML is limited. Eventually, they find it hard to implement. It's not because of the complexity but because of the resistivity. Learning new technology always requires patience, and that patience brings maturity.
We always encourage the stage of getting surrounded by myths because it shows that you are exploring the areas. But as true learners, it should be our responsibility to know the fact of every tale in any domain.
We may have cleared most of the misconceptions that are very popular for early learners.
Get well-designed application and interview centirc content on ds-algorithms, machine learning, system design and oops. Content will be delivered weekly.