Scribd
Upload
43 views

Introduction - To - ML Lec

IITM DS MLT lec

Uploaded by

DEBENDRA KUMAR DHIR
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
43 views

Introduction - To - ML Lec

IITM DS MLT lec

Uploaded by

DEBENDRA KUMAR DHIR
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 15

Machine

learning
Techniques
Lecture 1
Introduction

Instructor:
Arun Rajkumar

Image: Google
Top 10 Digital Transformation Trends

• 5G
• A faster WiFi
• Analytics
• AI and Machine Learning
• Blockchain
• RPA

Why? •
•

•
Conversational AI
Connected vehicles, autonomous drones and
Smart Cities
XaaS, UX/CX, and privacy

https://www.forbes.com/sites/danielnewman/2019/07/14/top-10-digital-
transformation-trends-for-2020/#39f7aef076be
Text

Brain
Vision

Where?
Speech

Image: Google
Procedural Data driven
Eg: Tax calculation

What?
Memorize Generalize

Magic Math!
Examples
• Spam vs Non-spam

• Forecasting rainfall

• Recommending movies

• Friend suggestions

• Voice/Instrument separation

• Grouping pictures in phone

• Robot navigation

• Stock market prediction

• etc …
Broad Paradigms

Supervised Learning Unsupervised Learning Sequential Learning

• Classification • Clustering • Online learning


Binary, Multiclass, Ordinal • Representation learning • Multi-armed Bandits
• Regression • Reinforcement learning
• Ranking
• Structure learning

Image: Google
Examples
• Spam vs Non-spam – Binary Classification

• Forecasting rainfall - Regression

• Recommending movies – Ordinal Classification

• Friend suggestions – Link Prediction

• Voice/Instrument separation – Unsupervised learning

• Grouping pictures in phone - Clustering

• Robot navigation – Reinforcement Learning

• Stock market prediction – Online learning

• etc …
Pre Requisites

Linear Algebra
Probability/Stats
Basic high school calculus
Goals
• Obtain a clear understanding of various paradigms of machine
learning

• Given a real world problem, develop ability to pose it as a relevant


machine learning problem.

• Understand key ML algorithms and their differences and


similarities, pros and cons.

• Learn the theoretical/statistical underpinnings of ML and how it


relates to practice.

• Appreciate the mathematics involved in the algorithms.

• Implement ML algorithms from scratch and demonstrate the


relevant output of interest using graphs and plots.
A Simple Example

200

Height
(in cms)

100

30 Weight 150
(in Kgs)
A Simple Example

200 STRUCTURE

Height
(in cms) UNCERTAINITY

DECISION

100

30 Weight 150
(in Kgs)
Foundations

STRUCTURE Linear Algebra

UNCERTAINITY Probability

DECISION Optimization
Tentative Roadmap
Unsupervised Learning
Representation Learning
Unsupervised Learning
Clustering

Unsupervised Learning
Estimation

Supervised Learning
Basic algorithms

Supervised Learning
Advanced algorithms
Supervised Learning
Advanced Topics

Image: Google
References
Welcome and
All the best!

You might also like