Scribd
Upload
239 views1 page

Assignment #1 Machine Learning

This document provides instructions for Assignment #1 in the BITS F464 Machine Learning course. It consists of two parts worth a total of 10% of the course grade. The first part involves performing linear regression on a dataset using different basis functions and comparing a Bayesian linear regression model to the other models. The second part tasks students with writing a program to implement a multilayer perceptron neural network for classification problems, allowing for different activation functions and gradient descent algorithms, and comparing the results. The assignment is due on November 9th, 2019 and should be submitted as a handwritten report. Groups of up to 3 students are permitted.

Uploaded by

SujeetSrivastava
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
239 views1 page

Assignment #1 Machine Learning

This document provides instructions for Assignment #1 in the BITS F464 Machine Learning course. It consists of two parts worth a total of 10% of the course grade. The first part involves performing linear regression on a dataset using different basis functions and comparing a Bayesian linear regression model to the other models. The second part tasks students with writing a program to implement a multilayer perceptron neural network for classification problems, allowing for different activation functions and gradient descent algorithms, and comparing the results. The assignment is due on November 9th, 2019 and should be submitted as a handwritten report. Groups of up to 3 students are permitted.

Uploaded by

SujeetSrivastava
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/ 1

BITS F464 - Machine Learning

I Semester 2019-2020
Assignment #1
Weightage: 10%
Due Data: 09th November, 2019 (Saturday)

1. Linear basis function models for Regression


(a) Take a regression problem for which data is available (check UCI ML Repository)
and try out different basis functions.
(b) Build a Bayesian Linear Regression Model* and compare it with models developed
in (a). DO a quantitative as well as qualitative comparison.

*Chapter 3 of Bishop Book on Pattern Recognition and Machine Learning. Bayesian Linear Regression is
the new model you will learn through this assignment.

You need to submit a hand-written report. Figures, tables, and graphs can be taken
printed.

2. Build your first Neural Network


(a) You are required to write a program for a Multilayer Perceptron (MLP) classifier in
C or Java (no Python, no R) with 1 hidden layer. You should provision for use of
different activation functions and different variants of Gradient Descent
algorithms.
(b) Generalize your code to work with any number of hidden layers.
(c) Test your code using one binary and one multiclass classification problem.
(d) Compare accuracies using different activation functions and different variants of
Gradient Descent. (using 1 hidden layer only)
Objectives: To understand the intricacies of Neural Networks and to get some insights into the working
of deep neural networks.

Group Information:
Max. 03 students are allowed per group. NO RELAXATION!

Navneet Goyal
goel@
6121-K, NAB.

You might also like