Ashish Kulkarni

Ashish Kulkarni

SDE at Nasdaq, Bachelor's degree in CS from PES University

Sitemap

A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.

Pages

Posts

portfolio

OpenGL: Physics Simulations and Cursor Interactions

A space for my OpenGL projects. It starts with basic 2-d projects, such as the Sierpinski triangle, or building a convex hull around points using Graham’s scan algorithm, all the way up to implementing elastic sphere collisions in 3-d space.

covibot

A Reddit bot which gives COVID-19 stats of a specific region to Redditors based on their post title.

Image Editor

Image editing command-line with implementations of greyscale, Gaussian blurring, masking based on colour range (HSV), etc. from scratch using OpenCV, NumPy, Pillow and Matplotlib.

Connect4

A multiplayer simulation of “Connect Four” by Hasbro using Python 3 socket programming.

Quantum Gate Simulator

PESU I/O: Simulates basic quantum gates such as the Hadamard gate or the Pauli-X gate on a system of two qubits.

publications

Diffusion Inference with Dynamic Classifier-free Guidance

Published in International Conference on Inventive Computing and Informatics, 2024

This research work proposes varying the CFG scale values across the inference steps by making use of various scheduling functions, which not only results in better images but also unlocks the full potential of the rich latent space representation of diffusion models by allowing for the sampling of various different images from the same initial conditions by only varying the CFG scale and keeping all other parameters constant.

Recommended citation: V. S., A., Kulkarni, A., Chawla, D., Rawther, A., & Rangareddy, J. (2024). Diffusion Inference with Dynamic Classifier-free Guidance. In International Conference on Inventive Computing and Informatics (2nd ed., pp. 53–59). IEEE. https://doi.org/10.1109/icici62254.2024.00018
Download Paper

A Comparative Study on Deep Convolutional Neural Networks and Histogram Equalization Techniques for Glaucoma Detection From Fundus Images

Published in medRxiv, 2024

This research aims to evaluate the performance of eight different convolutional neural network (CNN) models and histogram equalization techniques for glaucoma detection from fundus images. The study utilized the {ACRIMA} database, comprising 705 fundus images (396 glaucomatous and 309 normal), and data augmentation methods were employed.

Recommended citation: Kulkarni, A., & Ahmed, S. H. (2024). A Comparative Study on Deep Convolutional Neural Networks and Histogram Equalization Techniques for Glaucoma Detection From Fundus Images. Cold Spring Harbor Laboratory. medRxiv. https://doi.org/10.1101/2024.10.25.24316109
Download Paper

talks

Linear Regression Made Simple

Published:

Many people, especially those new to the field, often show great interest in the field of machine learning, but shy away from the mathematics that govern it. I hope to demystify it for you by explaining, in simple terms, the math behind machine learning.

GameStop: What it means to short a stock

Published:

This year, when a community on Reddit famously manipulated GameStop shares, I was happier than I’d ever been to see a fresh form of volatility in the markets. For those of you who are somehow oblivious to this incident, this article is for you. And those of you who’d like a deeper analysis into this incident, although you’ve heard of it, this article is for you too.

teaching

Undergraduate Teaching Assistant

Object Oriented Analysis and Design with Java, Department of Computer Science, PES University, 2024

During my 8th undergraduate semester, I served as a teaching assistant under Prof. Priya Badrinath, for 180+ students. My GitHub repository contains assignments and files I created during my stint as an undergraduate TA for the 6th-semester course Object-Oriented Analysis and Design with Java (UE21CS352B).

  • Created lab contests to test students’ understanding of Java and OOP concepts including inheritance, polymorphism, multi-threading, serialization, etc. Crafted a total of 14 detailed problem statements, generated test cases to evaluate them, and hosted them on HackerRank.
  • Wrote multi-page case studies for students to create use-case models, class diagrams, activity diagrams, and Java solutions.
  • Created a detailed guide on the model-view-controller architectural pattern, and wrote a case study for students to build a Java MVC solution to.
  • Solved and shared solutions for every task I assigned.