Simplifying the World of Machine learning & Deep Learning.
I, Rushi Prajapati, Welcome you to my another blog in my “Simplifying Series”, in which I try to explain complex topics by simplifying them. My first blog in this series was on computer vision, and now I’m bringing you another on the fascinating subject of ML-DL.
“Machine learning is the canvas, and deep learning is the brushstroke that adds depth and complexity to the artwork of AI”
AI uses a technique called ML (Machine Learning) to teach the computer how to learn from data and make predictions or decisions. ML is like giving the computer a set of rules to follow based on patterns it finds in the data. Deep Learning (DL) is a more advanced form of ML that mimics the human brain by using artificial neural networks., AI is like the big picture, ML is the technique used to teach the computer, and DL is a more advanced version of ML that makes the computer smarter.
Before exploring deep learning, let’s quickly understand machine learning.
Machine learning is a method of data analysis that uses algorithms that iteratively learn from data to find hidden insights/patterns automatically without being explicitly programmed.
- Each task requires a different set of algorithms, and these algorithms detect patterns to perform certain tasks.
- Like Human learning from past experiences.
- A computer systems learns from historical data, which represents some past experiences of an application domain.
- The goal is to learn a target function that can be used to predict the response variable (Regression/Classification).
WHAT IF I ASK YOU,IF YOU HAVE EVER UTILIZED MACHINE LEARNING?
- When you ask Siri what the weather forecast is, that’s machine learning.
- When you Google search something at work to help you do your job better or more efficiently, you can thank machine learning.
- Another everyday example is our spam folders — a machine learning algorithm is used to determine which emails are inbox-worthy, and which are spam and don’t deserve attention.
- Similarly, when Netflix suggests a show, you should watch based on preference, it’s getting the suggestion from an algorithm.
TYPES OF MACHINE LEARNING
Supervised Machine Learning
lets see with one simple example what’s Supervised Machine Learning
Imagine you have a basket filled with different types of fruits like apples, oranges, and bananas. You want to teach a machine to recognize these fruits. Here’s how it works:
- First, you show the machine a set of fruits and tell it what each fruit is called. For example, you show it an apple and say, “This is an apple,” then you show it an orange and say, “This is an orange,” and so on. This step is like teaching the machine with labeled examples.
- The machine pays close attention and learns from your explanations. It starts to recognize patterns and features that distinguish each fruit. It understands that apples are usually round and red, oranges are orange and have a bumpy texture, and bananas are yellow and have a curved shape.
- After the training is complete, you give the machine some new fruits that it hasn’t seen before. The machine examines these fruits and tries to guess what each one is based on what it learned during training.
- Using the knowledge it gained from the labeled examples, the machine makes predictions. For example, if you give it a round, red fruit, it will most likely say it’s an apple. If you provide it with a yellow, curved fruit, it will likely guess it’s a banana.
In this way, the machine learns to recognize different fruits by analyzing the labeled examples you provided during the training process. Supervised learning is like teaching the machine using labeled data, so it can make accurate predictions or identify things correctly when given new, unlabeled examples
I think you are understanding easily with the example, so let’s see another example and understand what unsupervised machine learning is.
Unsupervised Machine Learning
Imagine you have a collection of pictures that contain both dogs and cats, but you haven’t told the machine which pictures have dogs and which ones have cats. Now, the machine needs to figure out how to group these pictures on its own. Here’s how it works:
- The machine starts by looking at all the pictures and tries to find similarities, patterns, and differences among them. It analyzes things like shapes, colors, and textures.
- It notices that some pictures have animals with pointy ears, wagging tails, and a certain body shape, while other pictures have animals with round ears, purring sounds, and a different body shape.
- Based on these observations, the machine decides to group the pictures into two categories. In one group, it puts all the pictures that have animals with pointy ears, wagging tails, and a specific body shape. In the other group, it puts all the pictures that have animals with round ears and a different body shape.
- The machine has now created two groups without knowing anything about dogs or cats specifically(i.e clustring). It has simply noticed the similarities and differences between the pictures and grouped them accordingly.
In this way, the machine uses unsupervised learning to group the pictures based on the patterns it discovered. It doesn’t need any prior knowledge or labeled examples to do this. Unsupervised learning is like letting the machine explore the data on its own, finding similarities and differences, and organizing the information into meaningful groups without any specific guidance.
Importance of Machine learning
So, now that you are aware of machine learning, let’s explore deep learning.
Deep Learning is a subset of machine learning techniques that are based on artificial nueron network. It is a type of computation model inspired by the stucture and fucntioning of the human brain.
In Machine learning:
In Deep learning:
Deep learning skips the manual steps of extracting features, you can feed any kind of data to Deep learning model,which predicts the output.
Deep learning algorithms are designed to automatically learn and extract intricate patterns and representation from the large data, allowing the system to make accurate prediction, classification or decision.
The term “deep” in deep learning refers to the multiple layers of interconnect neurons or nodes that make up the neural network.
Deep learning has gain significant attention and success in various fields including Computer vision, Natural Language Processing, Speech Recognition and Robotics. Also in Sentimental analysis and Autonomous vehicles.
Deep learning is a very much inspired from the biological neuron and human brain.
A biological neuron receives electrical signals from its dendrites, modulates the electrical signals in various amounts, and then fires an output signal through its synapses only when the total strength of the
input signals exceeds a certain threshold. The output is then fed
to another neuron, and so forth.
Just like our brain has layers of interconnected cells called neurons, deep learning models also have layers of artificial neurons called artificial neural networks. These artificial neurons do math calculations on the input they receive. Each layer of neurons takes the output from the previous layer and does its own calculations on it. This creates a step-by-step process, where each layer learns to understand more complex things based on what the previous layer learned.
Deep learning is good at finding patterns in data. The first layers in the network look for simple patterns, like straight lines or basic shapes, while the later layers combine these simple patterns to recognize more complicated things, like objects or ideas.
This process of building up layers to understand complex things is similar to how our brain works. Our brain’s neurons start by recognizing simple things, like edges, and as we go higher up, they combine those edges to understand more complex things, like faces or animals.
Deep learning models are really good at tasks like recognizing images, understanding speech, and even playing games. They learn from a lot of examples to find and remember patterns, which helps them make accurate predictions or come up with new ideas.
It’s important to remember that deep learning models are not exactly like our brain. They’re simplified versions designed to learn and process information efficiently. But they do capture some important aspects of how our brain works and use them to do smart things with computers.
CONCLUTION
In conclusion, machine learning (ML) and deep learning (DL) have changed the way we do things in many industries. They help computers learn from lots of information to make smart predictions. ML and DL are like super tools that can recognize images, understand speech, and even talk to us like humans.
With each blog in my “Simplifying Series”, I want to make these complex topics easier to understand. I want you to feel confident exploring and learning about ML and DL.
Keep an eye out for more blogs in the “Simplifying Series.” I’ll keep explaining things in simple ways so that we can all learn and understand together. Let’s explore the world of technology and become knowledgeable about these amazing things.
