Srushti Patil

Hi there! I’m Srushti, currently a Research Assistant at IIT-Bombay, working on Many-body inspired quantum circuit compactification and associated error mitigation. I graduated a few months back from IISER Tirupati with a MS in Physics. My research interests are in the areas of Quantum Algorithms, Quantum ML, and Quantum Cryptography.

During my highschool years, I used to love solving Physics problems like finding the charge on a hollow sphere or the crazy relative velocity ones! When I first learned about Kepler’s Laws, I was awstruck by the very idea of planets following trajectories and how intact is everything, not too big, not too small, just perfect! And we can actually do some math about it!(There is a podcast by Lex Fridman with Leonard Susskind which touches the math part very beautifully) This led me to pursue my career in Physics as major durig my undergraduate years.

Working on a research problem; very similar to starting a band together :D

phdcomics

I am always open to discuss physics and any wild ideas that come to your mind! I love One Direction btw! .. and a bit of The Lumineers too!

Next stop: PhD?

news

Nov 3, 2023	Joined IIT-B as a Research Assistant
Jul 18, 2023	Graduated from IISER Tirupati with major in Physics.
Jun 1, 2023	Joined IBM Research as a research intern
May 2, 2023	Defended my thesis and got my first work arxived!

selected publications

Hybrid quantum-classical graph neural networks for tumor classification in digital pathology

Anupama Ray, Dhiraj Madan, Srushti Patil, and 2 more authors

2023

Abs

This paper presents a novel approach in the intersection of quantum computing and machine learning, specifically focusing on the application of a hybrid quantum-classical graph neural network (GNN) for tumor classification in digital pathology. The proposed methodology integrates a Variational Quantum Classifier (VQC) with GNN, exploring two variants: one with fixed pretrained GNN parameters and another with end-to-end training of GNN+VQC. The study reveals that the hybrid model achieves comparable performance to state-of-the-art classical GNNs in weighted precision, recall, and F1-score. Additionally, it demonstrates the efficacy of amplitude encoding in compressing information with a logarithmic number of qubits, outperforming classical compression techniques. The research also highlights the advantages of end-to-end training in enhancing the model’s performance.