Subhasish Basak

Principal Component Analysis and its application in Machine Learning

Spring 2019

As a part of my coursework at CMI, Numerical Linear Algebra & its applications : DG-1201, me along with my teammates prepared a project on PCA and its application. The objective of this project was to analyse how PCA can be used to find patterns to reduce the dimensions of the dataset with minimal loss of information.

• GitHub Repository
• |
• Project proposal

keywords/tools used:Singular Value Decomposition, Eigen-values & Eigen vectors, Python, Latex, GitHub

Applied Machine Learning Projects

Spring 2019

As a part of my post-graduate coursework at CMI, Applied Machine Learning I-II : CGE-118 & CGE-128 I completed several ML projects on Regression & Classification, Cluestering, Topic Modelling using LDA, Building movie recommendation system, Time series analysis and more. The course was focused on building basic foundation of ML concepts and was a great opportunity to learn ML algorithms implementation using python.

• GitHub Repository

Technologies Used: python, Latex, Jupyter notebook, Git, GitHub

Summer Internship at the Laboratoire des signaux et systemés(L2S)

École CentraleSupelec, Université Paris-Saclay

Summer 2019, May-July

I worked on Reviewing Scalability of Python toolboxes for Gaussian Process Regression under the supervision of Professor Emmanuel Vazquez (L2S, CentraleSupelec). As a part of my work, I built a universal wrapper for Python toolboxes implementing Gaussian Process regression, to perform comparative performance tests using simulated test beds and reviewed the toolboxes over different features, for future developement.

• GitHub Repository
• |
• report

Technologies Used: python, Spyder, Latex, git, GitLab

Credit Suisse top quant hackathon

Fall 2019

I participated in a team of 3 members from CMI in this inter college hackathon and our team finished among the top 3 selected teams from our institute. We addressed the problem of Anomaly detection in financial data. The objective was to develop innovative models to identify anomalies in an uncleaned dataset.

• GitHub Repository
• |
• Problem Statement

Tools used: Python, Google docs, git, GitHub

IFCAM Winter School on Graphs & Stochastic Process

Indian Institute of Science Education and Research, IISER - Pune

Winter 2019

This winter school was organised by the Indo-French Center for Applied Mathematics, here I attended mini courses on Branching random walks, Galton-Watson trees and random networks.

• Certificate
• |
• Site

Keywords : Stochastic processes, branching random walks, Galton-Watson trees, random networks