Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
12 views

Lecture 01 Introduction Annotated

This document provides an introduction to machine learning, including definitions and examples. It discusses supervised and unsupervised learning algorithms. The course will cover classification, regression, clustering and deep learning. Assignments include tutorials, quizzes, projects and exams.

Uploaded by

dikshant8005
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
12 views

Lecture 01 Introduction Annotated

This document provides an introduction to machine learning, including definitions and examples. It discusses supervised and unsupervised learning algorithms. The course will cover classification, regression, clustering and deep learning. Assignments include tutorials, quizzes, projects and exams.

Uploaded by

dikshant8005
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 18

Introduction to Machine Learning

Instructor: Prof. Ganesh Ramakrishnan


Lecture 1 : Introduction and Motivation
Introduction: What is Machine Learning?

Machine learning is a sub-field of computer science that


evolved from the study of pattern recognition and
computational learning theory in artificial intelligence.
In more simpler terms:
Using algorithms that iteratively learn from data,
Allowing computers to find hidden insights without being
explicitly programmed where to look
Introduction: What is Machine Learning?

Typical algorithm has a (large) number of parameters whose


values are learnt from the data
Applications include:
Hand Written digit recognition
Face Detection
Spam Detection
Speech recognition in Google Now
Real-time ads on web pages and mobile devices
....
Traditional Programming

Machine Learning
Example: Spam Detection
How to proceed...

This is an example of supervised learning problem:


data
training
testing
Example: Handwritten digit recognition

Figure: Digit recognition: Images are 28 ∗ 28 pixels

Represent input image as a vector x ∈ R 28∗28


Learn a classifier f (x) such that,
f : x → {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}
How to proceed...

This is an example of supervised learning problem:


data
training
testing
Course Overview
Supervised classification
perceptron
support vector machine
loss functions
kernels,
neural networks and deep learning

Supervised regression
linear regression
least square linear regression model
Bayes Linear Regression
non-linear regression
ridge regression
lasso regression
SVM regression

Unsupervised learning
clustering. K-Means
Expectation Maximization. Mixture of Gaussian
Logistics
Prerequisites
basic Linear Algebra
basic Probability Theory
huge interest in learning new algorithms
Tutorials
1 Tutorial sheet handed out every week, including a ‘Tutorial 0’
on the pre-requisites.
2 Expect students to try out each tutorial as homework
3 Solutions will be discussed at 1:30 PM before the following
class.
Assignments/Homework (Individual) -
2 assignments closely following content covered in class
Project - Group of 4
Divided into 3 stages
Stage 0 - Idea Proposals
Stage 1 - Initial report on data-sets etc
Stage 2 - Milestone
Stage 3 - Final Presentation
Quizzes
Quiz 1 - Week 3-4
Quiz 2 - Week 12
Midsem
Endsem

Assignments & Class Participation 20%


Quizzes 15%
Project 20%
Midsem 15%
Endsem 30%

Audit students have to attend classes, and submit assignments and


project.
Course Materials

Notes will be periodically posted at ‘cs725/calendar.html’ and on


moodle.

Primary Book:

Elements of Statistical Learning, Trevor Hastie, Robert


Tibshirani, Jerome Friedman, Springer
The following books are recommended for additional reading:
Pattern Recognition and Machine Learning, Christopher
Bishop, Springer, 2006.
excellent in classification and regression
Tom Mitchell, Machine Learning. McGraw-Hill, 1997
good explanation of algorithms and a bible for the course
Kevin Murphy, Statistical Machine Learning
Classroom Policy

Class Participation: Every student will get points based on


their participation in the following forms:
Homework questions
Class discussion, answering questions, asking good/foolish
questions
Piazza/Bodhitree participation for discussing Tutorial and
Specially Marked Questions (No private posts please!!)
Anything and everything which will make the course interesting
Honor Code

We want you to take a pledge that you will not be involved in any
sort of plagiarism.

All the assignments, projects and quizzes will be checked for copy
cases. In case of even a small case of copying, the name of both
the parties will be handed over to the DAC1

We also take a pledge that any sort of plagiarism will receive very
strict reactions2 .

1
http://www1.iitb.ac.in/newacadhome/punishments201521July.pdf
2
http://www1.iitb.ac.in/newacadhome/procedures201521July.pdf
Few Quotes

A breakthrough in machine learning would be worth ten


Microsofts - Bill Gates, Chairman, Microsoft
Machine learning is the next Internet - Tony Tether,
Director, DARPA
Machine learning is the hot new thing - John Hennessy,
President, Stanford
Web rankings today are mostly a matter of machine learning -
Prabhakar Raghavan, Dir. Research, Yahoo
Machine learning is going to result in a real revolution - Greg
Papadopoulos, CTO, Sun
Machine learning is today’s discontinuity - Jerry Yang, CEO,
Yahoo

You might also like