Scribd
Upload
58 views28 pages

Lectures1 2

This document provides an overview of the instructors and teaching assistants for a course on pattern recognition and machine learning. It introduces Richa Singh as the lead instructor and her background. It then lists the teaching assistants and their roles. Finally, it outlines the course objectives, learning outcomes, content, evaluation components, reading resources, and class organization.

Uploaded by

Rohit Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
58 views28 pages

Lectures1 2

This document provides an overview of the instructors and teaching assistants for a course on pattern recognition and machine learning. It introduces Richa Singh as the lead instructor and her background. It then lists the teaching assistants and their roles. Finally, it outlines the course objectives, learning outcomes, content, evaluation components, reading resources, and class organization.

Uploaded by

Rohit Singh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 28

Pattern Recognition and Machine

Learning
Richa Singh

Google classroom code: myiblie

Slides are prepared from several information sources on the web and books
About the instructors
• Richa Singh:
– Professor and HoD - CSE, IIT Jodhpur
– Faculty, IIIT Delhi (2009 – 2019)
– home.iitj.ac.in/~richa

– Adjunct Professor, West Virginia University


– Vice President – Publications, IEEE Biometrics Council
– Associate Editor-in-Chief, Pattern Recognition Journal
– Program Co-Chair, CVPR 2022
About the instructors
• Dr. Dipanjan Roy
– Associate Professor, AIDE School
Teaching Assistants
Debalina Saha Devin Garg
Garvit Chugh Harsh Rajiv Agarwal
Amit Kashalya Harshil Singh Bhatia
Nikhil Dwivedi Kartik Narayan
Neelu Verma Kwanit Gupta
Sandeep Khanna
Km Bhavna

BTech, MTech and PhD


students

• Added in the Google classroom


Course Objectives
• To understand various key paradigms for
pattern classification and machine learning
approaches
• To familiarize with the mathematical and
statistical techniques used in pattern
recognition and machine learning.
• To understand and differentiate among
various pattern recognition and machine
learning techniques.
Learning Outcomes
• The students are expected to have the ability
to:
– To formulate a machine learning problem
– Select an appropriate pattern analysis tool for
analyzing data in a given feature space.
– Apply pattern recognition and machine learning
techniques such as classification and feature
selection to practical applications and detect
patterns in the data.
Course Content
• Introduction: Definitions, Datasets for Pattern Recognition, Different Paradigms of
Pattern Recognition and Machine Learning, Data Normalization, Hypothesis
Evaluation, VC-Dimensions and Distribution, Bias Variance Tradeoff, Regression
(Linear) (8 Lectures)
• Discriminative Methods: Distance-based methods, Linear Discriminant Functions,
Decision Tree, Random Decision Forest and Boosting (5 Lectures)
• Bayes Decision Theory: Bayes decision rule, Minimum error rate classification,
Normal density and discriminant functions, Bayesian networks (7 Lectures)
• Parameter Estimation: Maximum Likelihood and Bayesian Parameter Estimation (3
Lectures)
• Feature Selection and Dimensionality Reduction: PCA, LDA, ICA, SFFS, SBFS (4
Lectures) Artificial Neural Networks: MLP, Backprop, and RBF-Net (4 Lectures)
• Kernel Machines: Kernel Tricks, Support Vector Machines (primal and dual forms),
K-SVR, K-PCA (6 Lectures)
• Clustering: k-means clustering, Gaussian Mixture Modeling, EM-algorithm (5
Lectures)
Reading Resources
• Textbooks:
– Pattern Classification, 2nd Edition, Richard O. Duda, Peter E.
Hart, David G. Stork, Wiley

• Reference Books:
– Tom Mitchell, Machine Learning
– C. Bishop, Pattern Recognition and Machine Learning, Springer
– K. Murphy, Machine Learning: a Probabilistic Perspective, MIT
Press
– Shalev-Shwartz,S., Ben-David,S., (2014), Understanding Machine
Learning: From Theory to Algorithms, Cambridge University
Press
Evaluation Components
• Grading
– Assignments (labs, programming and written): 30%
– Exams: 30%
– Project: 20%
– Quiz : 10%
– Kaggle Challenges: 10%
• Project team size: 2 students
– Predefined project topics: you have to select one
• Assignments: individually
Collaboration Policy
• Discussion with friends and colleagues is
good… but
– the objective should be to improve understanding
and learning
– Not getting answers
• If you have discussed with anyone, you should
acknowledge who helped you – from the class
or outside the class
Plagiarism Policy
• Cheating in assignments/quizzes/projects/
– First offence: Zero in the evaluation component
– Second offense: Grade reduction/F grade

• Cheating in exam: F grade

• Misbehavior: Institute guidelines


Class Organization

• Lectures: 3 days
• Labs: 2 groups
– Time will be decided this week
Any questions regarding
administrative guidelines?
What is Pattern Recognition and
Machine Learning?
Machine Learning
• What do we understand by learning?
– Learning is any process by which a system improves
performance from experience.”
– Herbert Simon (1950)
• Machine Learning is the study of algorithms that
– improve their performance P
– at some task T
– with experience E.
A well-defined learning task is given by <P, T, E>.

- Tom Mitchell (1998)


How to design a learning system
• Understand the problem statement
• Choose exactly what is to be learned
• Choose how to represent the target function
• Choose a learning algorithm to infer the target
function from the experience

Based on slide by Ray Mooney


Applications of ML

Weather forecast
What are the facial expressions?

What are these letters?


Are they same or different?
Applications of ML
Task, T
• Classification
• Regression
• Ranking
• Recommendation
• Clustering
• Density estimation
• ….
Performance, P
• Metric used to evaluate the performance of T
– Classification: error rate or accuracy
– Regression: mean squared error
– Density estimation: probability assigned to
samples
ML in Practice

• Understand domain, prior knowledge, and goals


• Data integration, selection, cleaning, pre-processing,
etc.
• Learn models
• Interpret results
• Consolidate and deploy discovered knowledge

Based on a slide by Pedro Domingos


Machine Learning Pipeline
• Three steps:
– Training
– Validation
– Testing
Machine Learning Pipeline

Database
(Input, Feature Learnt Model
Preprocessing Learning
Class labels) selection

Training
Machine Learning Pipeline

Validation

Evaluate the
Select the
Learning performance
Learning on
best model validation dataset

Database
(Input, Feature Learnt Model
Preprocessing Learning
Class labels) selection

Training
Machine Learning Pipeline
Testing

Test Input Feature


Preprocessing Classification Decision
extraction

Database
(Input, Feature Learnt Model
Preprocessing Learning
Class labels) selection

Training
Machine Learning Pipeline
Testing

Test Input Feature


Preprocessing Classification Decision
extraction

Database
(Input, Feature Learnt Model
Preprocessing Learning
Class labels) selection

Training
Typical Classification Algorithm
Typical Classification Algorithm

Performance Measure:
Accuracy (or fraction of correct answers) on test data

You might also like