00:00If you don't have a technical background but still want to learn the basics of artificial intelligence, stick around because we are distilling Google's 4-hour AI course for beginners into just 10 minutes

00:09I was initially very skeptical because I thought the course would be too conceptual

00:13We're all about practical tips on this channel, and knowing Google, the course might just disappear after 1 hour

00:18But I found that the underlying concepts actually made me better at using tools like ChatGPT and Google Bard, and cleared up a bunch of misconceptions I didn't know I had about AI, machine learning, and large language models

00:33So, starting with the broadest possible question: what is artificial intelligence? It turns out..

00:37I'm so embarrassed to admit I didn't know this, but AI is an entire field of study, like physics, and machine learning is a subfield of AI, much like how thermodynamics is a subfield of physics

00:51Going down another level, deep learning is a subset of machine learning, and deep learning models can be further broken down into something called discriminative models and generative models

01:01Large language models (LLMs) also fall under deep learning, and right at the intersection between generative models and LLMs is the technology that powers the applications we're all familiar with, such as ChatGPT and Google Bard

01:13Let me know in the comments if this was news to you as well

01:17Now that we have an understanding of the overall landscape and you see how the Different disciplines sit in relation to each other

01:22Let's go over the key takeaways you should know for each level, in a nutshell

01:27Machine learning is a program that uses input data to train a model, and that trained model can then make predictions based on data it has never seen before

01:38For example, if you train a model based on Nike sales data, you can then use that model to predict how well a new shoe from Adidas would sell based on Adidas sales data

01:47Two of the most common types of machine learning models are supervised and unsupervised learning models

01:54The key difference between the two is that supervised models use labeled data, and unsupervised models use unlabeled data

02:01In this supervised example, we have historical data points

02:05That plot shows the total bill amount at a restaurant against the tip amount

02:10Here, the data is labeled: blue dots indicate the order was picked up, and yellow dots indicate the order was delivered

02:15Using a supervised learning model, we can now predict how much tip we can expect for the next order, given the bill amount and whether it is picked up or delivered

02:24For unsupervised learning models, we look at the raw data and see if it naturally falls into groups

02:30In this example, we plotted employee tenure at a company against their income

02:35We see this group of employees has a relatively high income-to-years-worked ratio, versus this other group

02:42We can also see that all these are unlabeled data; if they were labeled, we would see male or female, and years worked

02:48Company function, etc

02:55We can now ask this unsupervised learning model to solve a problem like, if a new employee joins, are they on the FastTrack or not? If they appear on the left, then yes; if they appear on the right, then no

03:01Pro tip: another big difference between the two models is that after a supervised learning model makes a prediction, it will compare that prediction to the training data used to train that model, and if there's a difference, it tries to close that gap

03:16Unsupervised learning models do not do this, by the way

03:18This video is not sponsored, but it is supported by those of you who subscribe to my paid productivity newsletter on Google tips

03:25Link in the description if you want to learn more

03:27Now we have a basic graph as of..

03:31Machine learning

03:34It's a good time to talk about deep learning, which is just a type of machine learning that uses something called artificial neural networks

03:37Don't worry, all you have to know for now is that artificial neural networks are inspired by the human brain and look something like this: layers of nodes and neurons

03:46The more layers there are, the more powerful the model

03:51And because we have these neural networks, we can now do something called semi-supervised learning, whereby a deep learning model is trained on a small amount of labeled data and a large amount of unlabeled data

04:02For example, a bank might use deep learning models to detect fraud

04:06The bank spends a bit of time to tag or label 5% of transactions as either

04:11Fraudulent or not fraudulent, and they leave the remaining 95% of transactions unlabeled because they don't have the time or resources to label every transaction

04:19The magic happens when the deep learning model uses the 5% of labeled data to learn the basic concepts of the task

04:26Okay, these transactions are good and these are bad

04:30Okay, apply those learnings to the remaining 95% of unlabeled data, and using this new aggregate data set, the model makes predictions for future transactions

04:40That's pretty cool, and we're not done because deep learning can be divided into two types: discriminative and generative models

04:46Discriminative models learn from the relationship between labels and data points and only have the ability to..

04:54Classify those data points as fraud or not fraud

04:56For example, you have a bunch of pictures or data points, and you purposefully label some of them as cats and some as dogs

05:03A discriminative model will learn from the label "cat" or "dog," and if you submit a picture of a dog, it will predict the label for that new data point as "dog." We finally get to generative AI

05:16Unlike discriminative models, generative models learn about the patterns in the training data

05:22Then, after they receive some input— for example, a text prompt from us—they generate something new based on the patterns they just learned

05:28Going back to the animal example, the pictures or data points are not labeled as "cat" or "dog," so a generative model will look for patterns— oh, these data..

05:37Points, all have two ears, four legs, a tail, like dog food, and bark when asked to generate something called a dog

05:45The generative model generates a completely new image based on the patterns it just learned

05:51There's a super simple way to determine if something is generative AI or not

05:55If the output is a number, a classification, spam or not spam, or a probability, it is not generative AI

06:03It is general AI

06:10When the output is natural language text, or speech, an image, or audio, basically, generative AI generates new samples that are similar to the data it was trained on

06:18Moving on to different generative AI model types, most of us are familiar with text-to-text models like ChatGPT and Google Bard

06:25Other common model types include text-to-image models

06:27Like Midjourney and Stable Diffusion, these can not only generate images but also edit images as well

06:36Text-to-video models, surprise surprise, can generate and edit video footage

06:40Examples include Google's Imagen Video, Cog Video, and the very creatively named Make-a-Video

06:44Text-to-3D models are used to create game assets, and a little known example would be OpenAI's Point-E model

06:51Finally, text-to-task models are trained to perform specific tasks; for example, if you type "Gmail, summarize my unread emails," Google Bard will look through your inbox and summarize your unread emails

07:04Moving over to large language models, don't forget that LLMs are also a subset of deep learning, and although there is some overlap, LLMs and GNNs (Graph Neural Networks) are not the same

07:13The same thing, an important distinction is that large language models are generally pre-trained with a very large set of data and then fine-tuned for specific purposes

07:26What does that mean? Imagine you have a pet dog

07:28It can be pre-trained with basic commands like sit, come, down, and stay

07:33It's a good boy and a generalist

07:35But if that same good boy goes on to become a police dog, a guide dog, or a hunting dog, they need to receive specific training, so they're fine-tuned for that specialist role

07:45A similar idea applies to large language models

07:48They're first pre-trained to solve common language problems like text classification, question answering, document summarization, and text generation, then using smaller industry-specific datasets

08:02Datasets, these large language models are fine-tuned to solve specific problems in retail, finance, healthcare, entertainment, and other fields

08:11In the real world, this might mean a hospital uses a pre-trained large language model from one of the big tech companies and fine-tunes that model with its own first-party medical data to improve diagnostic accuracy from X-rays and other medical tests

08:25This is a win-win scenario because large companies can spend billions developing general-purpose large language models, then sell those LLMs to smaller institutions like retail companies, banks, hospitals, which don't have the resources to develop their own large language models but do have domain-specific datasets to fine-tune them

08:34Models

08:42Pro tip: if you do end up taking the full course, I'll link it down below

08:48It's completely free

08:54When you're taking notes, you can right-click on the video player and copy the video URL at the current time, so you can quickly navigate back to that specific part of the video

09:00There are five modules in total, and you get a badge after completing each module

09:04The content overall is a bit more on the theoretical side, so you definitely want to check out this video on how to master prompting

09:10Next, see you in the next video

09:12In the meantime, have a great one.