Welcome to the AI-ML Workshop
Decrypting the AI-ML hype and how artificial intelligence is reshaping the World!
Series of workshop hosted by TechConverge, SVNIT.
What is Machine Learning?
* * *
Definitions of ML
-
Field of study that gives computers the ability to learn without being explicitly programmed
Definitions of ML
-
"Field of study that gives computers the ability to learn without being explicitly programmed."
Arthur Samuel -
"A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E."
Tom Mitchell
Ans: Data

When Big Data meets Machine Learning the models flourish!!
-
Data is cheap nowadays at least relatively; Knowledge is harder to come by.
-
From data we are able to derive value from large, heterogeneous and disparate sources of data at lightning speed and scale.
-
ML helps to turn data into knowledge.
-
Spoiler Alerts
Machine learning is not the answer to all problems.
-
Limitation 1 — Ethics:The idea of trusting data and algorithms more than our own judgment has its pros and cons.
-
Limitation 2 — Data:Manifest itself in two ways: lack of data, and lack of good data.
-
Limitation 3 —Interpretability:If you cannot convince your client that you understand how the algorithm came to the decision it did, how likely are they to trust you and your expertise?
Confusing data

Type of Machine Learning
This is how machine learn Supervised!!?
-
Train
-
Test
-
Validation
.png)
This is how machine learn Supervised!!? (Formal way)
Given (X, Y), find the mapping between Y → X
We wanted to predict price of a house, give some features of a house.
edited.png)
This is how machine learn Supervised!!? (Formal way)
Given (X, Y), find the mapping between Y → X
We wanted to predict price of a house, give some features of a house.
With the help of this line we can predict Y given X
Here Y is a label
Regression vs. Classification
edited.png)
Regression
Prediction output belongs to the set of Real No.
Prediction output is continuous in nature.
Regression vs. Classification

Regression vs. Classification

Classification
Prediction output belongs to the set of discrete No. like here we have two classes. output will be either 1 or 2
Prediction output is a class.
Regression vs. Classification


The Unsupervised way
Given an Unlabeled data , Algorithm should make groups out of them.
For ex: Social media recommendation , Netflix recommendation

Reinforcement Learning
Model which make intelligenent decisions
for each decision Reward or penalty is assigned to agent
AI is trained to maximise the reward
Prerequisites
These are some prerequisites from our side.
- Basic Python: basic knowledge of python as well as list,dict,loops and if-else block
- Python Libraries: numpy,pandas,matplotlib
- Linear Algebra:understanding of matrix from 12th std.
- Basic Calculus: understanding of differenciation from 12th std.
- Probability and Statistics: bayesian theorem , mean,std. deviation,