This article was published as part of the Data Science Blogathon.
Humans should be concerned about the threat posed by artificial intelligence. – Bill Gates
I am sure that the above quote wants to convey a message to us, we should definitely think about it. What you think “I'm right?”, please share your opinion in the comment box, I will definitely read them, which will help me understand that “Is there any negative impact of these technologies on human species or not?” O "Will this technology be responsible for the extinction of the human species??”.
Let us begin, today our agenda is that we are going to discuss "Is there really a need for deep learning?”.
In the last few years we have probably heard a lot about deep learning, but what is it really about? Here's another curious question that comes to mind: “Why is deep learning only now taking center stage??”. Let's first understand what Artificial Intelligence really is.
Artificial intelligence
Artificial intelligence is a generic term for a branch of computer science. Its goal is for the machine to mimic human cognition, focusing on solving complex problems. The only goal of AI is that the Machine can have human-like intelligence in the future. It refers to the simulation of human intelligence in machines that are programmed so that the machine is able to think like humans and imitate their actions.
Machine learning is a subset of artificial intelligence that basically focuses on how to make a computer capable of learning on its own without the need for hand-coded instructions. Machine learning systems analyze a large amount of data and learn from past mistakes. The results are generated from algorithms that complete their task efficiently.
Deep learning is a subset within machine learning, this technology tries to imitate the activity of neurons in the human brain by multiplying matrices. This arrangement is called a neural network.. In reality, the concept of neural networks entered the scene in 1957 and was first tested in 1980, but it was not useful. The deep web only becomes feasible for just two reasons, the first is an increase in computing power and the second is a large amount of data. After reading up to this point, you will definitely doubt that whenever we talk about Deep Learning every time "big data" this term comes with Deep Learning. In reality, Why do neural networks need so much data?
The answer to the above question is, in reality, the more data, the more robust your network will be. Due to its robustness, your network will give better and more accurate results than any other algorithm. Let's put ourselves in the shoes of Deep Learning😁. Suppose you have seen 3 cat pictures, taken from different angles. But on the other hand, you have seen thousands of different cats, now it is much easier for you to recognize one. This is important for the data. En Deep Learning, data is the essence that allows the machine to learn.
Birth of the neural network
The true field of Deep Learning began in 2012, before 2012, most experts believed that the Neural Network was useless. In 2012, Deep Learning becomes the center of attention. In 2012, the neural network was first used in competition to recognize the world's largest image data set and, in fact, outperformed all previous algorithm types. In this movement, the world realizes the real power of neural networks. This was the birth of the Neural Network.
My opinion about “Is there a need for deep learning🤔?”
Before my opinion, let's see a graph researched by google.
According to Google search after 2013, most of the people did a lot of research on Deep Learning. The above graph shows the interest of people in the field of Deep Learning. The graph has been increasing long after 2013, then, What has led to this rise in trend that we will understand in this article?
According to some experts, there are many reasons behind this trend or the exponential growth of people's interest in the field of Deep Learning. Let's see them one by one.
1. The first thing is that after 2013 most people know about smartphones and start using them. People start using various social media platforms like Facebook, Instagram or WhatsApp, that actually generate a lot of data. By using this large amount of data, we can definitely do a lot of things, solve different types of use cases. Ex. recommendation system and many things.
Data is the top specific reason deep learning comes into the picture. According to a survey every day approx. 2,5 trillion bytes of data generated. Let's look at a beautiful graphic shared by Sir Andrew Ng.
Up here, we can see this specific graph where on the x-axis we have the amount of data and on the y-axis we have the performance of the algorithm. As we see, as we increase the amount of data over older learning algorithms (any type of machine learning algorithm), performance after a specific point of time, began to degrade and remains almost constant, did not increase. But in the case of Deep Learning, as we increase the amount of data, performance also increases. It means that this exponential data growth led us to create some amazing deep learning models in terms of accuracy and various performance metrics..
2. Technology is another reason that encourages us to research deep learning because, along with a large amount of data, deep learning also required good quality hardware. Here I am talking about GPU (Graphics processing unit) y TPU (Tensor Processing Unit). Due to the improvement of technology, now we easily get good hardware at a much lower price. As technology increases day by day, hardware cost drastically decreases day by day.
3. In reality, deep learning combines feature extraction and training part of the model. We carry out these two techniques separately in the case of Machine Learning, but here both techniques are included within the deep learning techniques.. Here feature extraction and model building, that we do it separately in the case of Machine Learning, are fully combined in Deep Learning projects. Because of this, Deep Learning can really solve complex problems like image classification, object detection or NLP task. Deep learning actually uses the deep neural network, as the neural network becomes deep and increasingly complex information and characteristics are extracted within a problem statement.
Let's finish😅!
In the previous points, we have discussed that technology supports us continuously, so why not take his hand and step forward? But this technology has a set of significant disadvantages despite all its benefits.. According to my, the deep learning model is unable to provide arguments as to why it reached a certain conclusion. I think it can cause some problems and can be challenging for the deep learning model. It is okay that it takes a lot of time and good hardware to train the model.
I think deep learning models should also give a specific conclusion about its output, suppose that every time someone asks us what “it is a cat?”, The way we use to explain why it's a cat, I think that the Deep Learning Model should also devise the same strategy that whenever it gives some result it will also give us an adequate conclusion. I don't think it's possible or not, I'm just sharing my views😅.
And my final opinion is that, in reality, we should keep going with the trends and technologies, that in fact help us to stay updated and grow more. What you think, leave a comment below.
Final notes!
Hope you enjoyed this article. Any question? Have i missed something? Please contact me at my LinkedIn. And finally, … No need to say,
Thank you for reading!
See ya!
The media shown in this article is not the property of DataPeaker and is used at the author's discretion.
