AI 101: AI Basics and the Role of Machine Learning

NOTE: You can listen to this content in this podcast also

AI 101: Understanding AI and the Role of Machine Learning in AI

Hello, everyone. This is Rabi Jay here again, and today, we will discuss an essential topic, Artificial Intelligence 101.?

I was initially surprised when one of my friends proposed that I do this topic, but then it became a relevant question. In fact, people have been asking me the difference between data analytics, AI, and so on. If we ask this question two or three years from now, it's like asking someone what a smartphone is, but it still makes sense today because I believe we are still in the initial stages of AI adoption. Moreover, this is such a rapidly moving space, and the definition keeps expanding.

The Next Step in Human Evolution

Artificial intelligence is a collection of technologies enabling us to mimic the human brain or, I should say, human intelligence. The dictionary definition of intelligence is "the ability to acquire knowledge and skills and apply them.".

Here's the definition from Wikipedia: Human intelligence is the?intellectual capability of humans, which is marked by complex cognitive feats and high levels of motivation and self-awareness.. Through intelligence, humans possess the cognitive abilities to learn, form concepts, understand, apply logic, and reason, including the capacities to recognize patterns, plan, innovate, solve problems, make decisions, retain information.

That's really what happens. Just look at that definition, folks—wouldn't you agree that AI is almost there? Other than emotional aspects such as motivation, AI can pretty much do everything out there. Based on that alone I would be courageous enough to say this below

Think of AI as Intelligence captured in a machine.

That comes from me, folks, and I know some purists may not like me saying that, but in essence, that's what it is trying to do??.

What is Artificial Intelligence?

Anything that a machine does, for example, listening to someone and responding back in human words? Take Alexa, for example, that is artificial intelligence, or when a machine looks at the surroundings using computer vision and then helps the car to navigate, which is also artificial intelligence. The technologies involve all those radars and sensors, devices that help us make all this happen, and devices that help us take in all the data and feed it into the car so that it can take appropriate action. That collection of technologies, such as natural language processing and computer vision, is what is known as artificial intelligence.?

Here's Gartner's definition of AI

Artificial intelligence (AI) applies advanced analysis and logic-based techniques, including machine learning, to interpret events, support and automate decisions and take action.

Machine Learning: The Brain behind Artificial Intelligence

But then, the other question is, what is machine learning? Because machine learning is often used interchangeably with artificial intelligence, the best way to answer that question is to take the driverless car example; Again, we see all the sensors and radars and everything, and as a whole ecosystem, the systems in the car can take in all the data from the environment and then move the car around. But then, how is that happening?

Understanding the Role of Machine Learning in AI

There is a machine learning model that takes in the data and then predicts the outcome that is required. So it tells the car to go left, right, stop, move fast, and accelerate.

Machine learning is about crunching all the data, processing it, identifying the patterns, and based on that, returning a prediction.

So that is the difference. Artificial intelligence is an umbrella term for all the technologies that try to mimic human intelligence. But the actual process by which this happens is through machine learning.?

How Machine Learning Overcomes the Limitations of Traditional Programming

Now, here's the other important, beautiful thing. Machine learning is different from traditional software programs. It is an entirely different programming paradigm. Because in the case of traditional programming, especially for me, I come from the SAP world. My first IT job was with SAP many years ago, and I am a great fan of SAP. There's only so much we can do with rule-based systems when considering product recommendations for the website. For example, there are so many permutations and combinations that you can't even write down the rules. There are millions of products and customers, each with unique attributes and behaviors; how do you capture all these rules??

Machine Learning relieves us from the shackles of rule-based programming

The Human-Like Approach to Defining Rules in Machine Learning?

As you know, we need to define rules for situations such as when this person comes and buys a product or views a movie on Netflix. We need to code rules that say, they are likely to buy these five other products, or because this customer is buying it, the other customers are also likely to buy it, or because this product was sold, other similar products will cross-sell or upsell. It's impossible to track all those rules.

That is the beauty of machine learning, because it tries to learn as a human does, which is based on analyzing incoming data.

Think of a baby; for example, when the baby learns to walk, the baby is making a lot of errors. The baby is experimenting. But eventually, the baby learns how to walk. The same happens when we learn how to drive a car for the first time or try to ride a bicycle; we make many mistakes. Eventually, we learn.?

How Data Scientists Train Machine Learning Models to Improve Performance?

That is what machine learning does: the data scientists try to train the model, tune it to perform better, but the models make many mistakes. Still, eventually, they learn, and they get better and better. So that is a fantastic invention in the field of artificial intelligence, which is making huge waves now.

This new paradigm of Machine Learning Algorithms has enabled speed and scale opening up new use cases that were not possible before!

And there is also the other field called deep learning and the transformer model, which is the basis of ChatGPT. We will talk about those in the following few articles.

I hope you liked this post. Please follow me on LinkedIn, subscribe to my?Enterprise AI Transformation Newsletter, and click the "like" button below. And share this post within your network and ask them to subscribe to this newsletter so more people can benefit.

Here's to?Enterprise AI Transformation, thanks!

Disclaimer: All opinions are my own. I am speaking for myself only and do not reflect the views of my employer.

?#Analytics??#ArtificialIntelligence??#aws?#BigData?#business?#BusinessIntelligence???#DataAnalytics??#DataManagement??#DataScience?#DataStrategy?#DataTransformations???#deloitte?#deloitteconsulting?#deloittedigital??#DigitalTransformation?#enterpriseai???#futureofAI?#AI???#digitalmarketing?#gartner?#forrester??#gartnerit??#Innovation??#MachineLearning?#mckinsey??#ml?#mlengineer??#sagemaker?#sap?#transformation??