BITS F464 Machine Learning Semester 1 2024-25

Birla Institute of Technology and Science - Pilani K K Birla Goa Campus Page 1 of 4

Handout BITS F464 Machine Learning

• Course Title: Machine Learning

• Course Number: BITS F464

• Instructor-in-charge: Aditya Challa, Ashwin Srinivasan

• Office Number: D245

1 Prerequisites
This course assumes familiarity with fundamentals of Linear Algebra, Probability and Statistics. We
also assume decent background in fundamentals of data science.

2 Overview
Data-driven computation is now a ubiquitous feature of a variety of programs designed for estimation,
modeling and inference. While the use of data to compute and rank answers to queries in search
engines like Google is well-known, large quantities of data are now available in areas that span
industry, sciences (both natural and social), and even the humanities.
The greater availability of data has brought to prominence computational tools that construct
models to compress, explain, predict, and simulate the data. A large class of such tools come
under the category of “machine learning”. Originally conceived as a necessary component for the
development of intelligent systems, machine learning is now a thriving scientific discipline in its own
right. This course will describe some of the tools, techniques and theory of machine learning

3 Objectives:
By the end of the subject, students should be able to:

• Set up a well-defined learning problem for a given task

• Select and define a representation for data to be used as input to a machine learning algorithm

• Select and define a representation for the model to be output by a machine learning algorithm

• Compare different algorithms according to the properties of their inputs and outputs

Prepared by Aditya Challa, Ashwin Srinivasan on August 28, 2024

Handout BITS F464 Machine Learning Page 2 of 4

• Compare different algorithms in terms of similarities and differences in the computational

methods used

• Describe some key algorithms to solve a learning problem in terms of the inputs, outputs and
computational methods used

• Use algorithms in applications to real-world data sets and

• Collect results to enable evaluation and comparison of their performance

4 Course Plan
• (Weeks 1) Preliminaries: (Aditya Challa/Ashwin Srinivasan)

• (Weeks 2-3) Statistical Learning Theory: Basics of Statistical Learning, Assumptions, Bias
Variance Tradeoff, Bayes Risk, Empirical Risk Minimization (Aditya Challa)

• (Weeks 4) Assessment of ML algorithms: Generalization Error, Estimation of Error, (Aditya

Challa)

• (Weeks 5-7) ML as optimization: Support Vector Machines, Neural Networks (Aditya Challa)

• (Weeks 8-10) Discriminative Models: Conditional probability models, classification and re-
gression (Ashwin Srinivasan)

• (Weeks 11-12) Generative Models: Joint probability models, clustering (Ashwin Srinivasan)

5 Text Book and References

The textbook will be followed throughout the course.

1. A First Course in Machine Learning, S. Rogers and M. Girolami, CRC Press, 2nd Ed (2015).
You should attempt to understand completely Chapters 1-6 of this book. The 1st edition of
this book is also OK, provided you match up the topics.

2. Hastie, T., Tibshirani, R., Friedman, J. H., & Friedman, J. H. (2009). The elements of
statistical learning: data mining, inference, and prediction (Vol. 2, pp. 1-758). New York:
springer.

3. Fernandes de Mello, R., & Antonelli Ponti, M. (2018). Machine learning: a practical approach
on the statistical learning theory. Cham: Springer International Publishing.
Handout BITS F464 Machine Learning Page 3 of 4

4. An Introduction to Statistical Learning, Gareth James, Daniela Witten,Trevor Hastie,Rob Tib-

shirani, Jonathan Taylor. Freely available at https://statlearning.com/. Also includes
some basic tutorials/labs in python.

References:

1. The Hundred Page Machine Learning Book, Andriy Burkov, 2019.

2. Data Mining, I.H. Witten and E. Frank, McGraw-Hill, 2011

3. Machine Learning, T. Mitchell, McGraw-Hill, 1997.

Popular Science:

1. The Master Algorithm, P. Domingos, Basic, 2015.

2. Probably Approximately Correct, L. Valiant, Perseus, 2012.

6 Evaluation
In-Class Quiz Project Mid-semester Examination Comprehensive Examination Total
2 × 10 = 20 10 30 40 100

• The in-class quizzes will be conducted in the CC Lab during the class/tutorial hours. The exact
dates will be confirmed within the 1 week from the start of the course.

7 Make-Up Policy
• There is strictly no makeup for the quizzes/labs.

• Makeup examinations for mid-semester/compre will only be given to students based on the
information we receive from AUGSD.

8 Important Information:
Assumed Knowledge Details of the minimum assumed knowledge for this course will be provided.
You are expected to know these topics. Some parts of these will be covered in the “Foundations”
Module of the course.
Time Commitment: Besides the lectures, you must be prepared to attend Tutorials and Labs.
This course will take about 8-10 hours of your time every week.
Handout BITS F464 Machine Learning Page 4 of 4

Scope: The lectures will not cover: Data Mining, Information Retrieval, Collaborative Filtering,
Social Media Analysis, Deep Learning and Reinforcement Learning. Some of these may be covered
in the labs.
Content: The lectures and tutorials are intended to give you the conceptual basis for some well-
known tasks in Machine Learning. The labs are intended to give you the opportunity to experiment
with popular tools and techniques. The labs should help you with your major and minor projects.
Delivery: All material will be made available on-line. If appropriate, some portion of the course
may run in a “flipped” mode. That is: you will be expected to review the material for the week
before attending the lecture, tutorial or lab. There will be a dedicated “office hour” every week for
clarification of doubts arising from the material covered up to that week. Random quizzes may be
used to assess understanding.
Plagiarism: It is expected that you are aware of the University’s definitions, policies and actions
on plagiarism All work submitted for assessment must be your own work. Copying is unacceptable.
Penalties for copying range from receiving no marks, to receiving an “NC” for the course and being
referred to the appropriate Institutional Board for academic misconduct.
All of the following are unacceptable: (a) Allowing someone to copy your work; (b) Submission
of work derived from another person, without their consent; (c) Submission of work derived from
another person with their knowledge; (d) Jointly written work, unless group submissions are explicitly
allowed; (e) Obtaining solutions for assessed components from external sources other than the ones
explicitly allowed.

9 Course Materials/ Notices/ Information

Course materials/notices/information will be made available through the google classroom. It is the
responsibility of the student to regularly check the google classroom for updates.

10 Office Hours
Monday, Wednesday 11:00 AM - 12:00 Noon.