Let's see how normies entertain themselves

Netflix and Chill

Chatting with friends

Artiste

How chad engineers entertain themselves

AI Art

Pissing off other gamers

We have all been there  😔

Fooling around with LLM's

What made all these examples possible ?

But what is Machine Learning ?

But what is intelligence ?

What is Artificial Intelligence ?

In other words, it's a process of training a computer system to identify patterns in data, and make predictions or decisions based on those patterns.

Patterns

Predictions

Some common misconceptions around ML

Fields in Machine Learning

Tabular Data

Applications of Tabular Data

Bussiness Data Analysis

Census

Computer Vision

Applications of Computer Vision

Face Recognition

Self-Driving Vehicles

Google Lens

Natural Language Processing

Applications of NLP

Hate Speech Detection

Chat Bots

Writing Assistance

Time-series

Applications of Time-Series

Stock Prediction

Weather Forecasting

• Regression

•  Regression

Used to predict a continous outcome

•  Regression

Used to predict a continous outcome

Example : Predicting your end-semester marks

•  Classification

•  Classification

Used to predict the label/class of an object

•  Classification

Used to predict the label/class of an object

Example : Predicting whether you pass your end-sems or not

Supervised Learning

But what exactly is labelled data ?

Let's take an interesting use-case where we try to predict the amount of likes of a Youtube video

Is it raining ?

Is it raining ?

Yes

Not going

Is it raining ?

Yes

Not going

Labs today ?

No

Is it raining ?

Yes

Not going

Labs today ?

No

No

Not going

Is it raining ?

Yes

Not going

Labs today ?

No

No

Not going

Yes

Is it raining ?

Mood hai ?

Is it raining ?

Yes

Not going

Labs today ?

No

No

Not going

Yes

Is it raining ?

Mood hai ?

No

Not going

Is it raining ?

Yes

Not going

Labs today ?

No

No

Not going

Yes

Is it raining ?

Mood hai ?

No

Not going

Yes

Going

Unsupervised Learning

• Unsupervised machine learning is a type of machine learning where the model is trained on unlabeled data, meaning there are no pre-defined labels or categories for the data.

• The goal of unsupervised learning is to find patterns, structures, or relationships within the data without being given any specific target to predict.

Applications of Unsupervised Learning     

Customer Segmentation

Customer Segmentation

In Search Engines                 

In Search Engines                 

Spam Filters

Spam Filters

Clustering

In clustering, we do not have a target to predict. We look at the data, try to club similar observations, and form different groups.

The most common example of clustering is the K-Means Clustering Algorithm.   

The most common example of clustering is the K-Means Clustering Algorithm.   

Reinforcement Learning

ML is great but here are some of it's shortcomings

•  Can't Handle Larger and Complex Real Life Data

Audio Data

• Scalability

• Doesn't Achieve High Accuracy

Suppose you have a problem

How will you create an image recognition program using ML?

Here’s a suggestion ….

Suppose you have a problem

How will you create an image recognition program using ML?

Here’s a suggestion ….

• Data Collection and Preprocessing

Suppose you have a problem

How will you create an image recognition program using ML?

Here’s a suggestion ….

• Data Collection and Preprocessing
• Feature Extraction and Selection

Suppose you have a problem

How will you create an image recognition program using ML?

Here’s a suggestion ….

• Data Collection and Preprocessing
• Feature Extraction and Selection
• Creating a model using SVMs/Logistic Regression/Random Forests or Naive Baiyes

Suppose you have a problem

How will you create an image recognition program using ML?

Here’s a suggestion ….

• Data Collection and Preprocessing
• Feature Extraction and Selection
• Creating a model using SVMs/Logistic Regression/Random Forests or Naive Baiyes
•  Model Evaluation 

Sounds like a handful, doesn’t it?

What if there is an easier way around all this……

What if there exists a way that combines three of the above steps into one…..

What a life saver, isn’t it?

But does something like that exist?

Yes … They’re called Neural networks….sounds familiar doesn’t it?

Most of you would already know that these “Neural Networks” relate to the human brain, the thing that makes our brain so special.

Well, something like that also exists in Machine learning, a “little” subset called Deep learning!

How does the Brain work?

The Basic functional unit of our Brain is Neuron

How Do we imitate the human brain?