Research Interests
My interests lie at the intersection of computer science and mathematics. I'm especially interested in the following topics:
- Graph Theory (algorithms and novel applications)
- Machine Learning (deep supervised learning, reinforcement learning)
- Music Information Retrieval (mostly music generation)
- Cryptography (especially graph applications)
Research Projects
Abstract: Music Information Retrieval (MIR) is a broad field with various goals such as genre classificaiton,
artist classification, and music generation. A common challenge in MIR is how to represent and
analyze music digitally. Common representations include digital audio formats, like .mp3 and .wav,
and MIDI, which contains more precise note information but lacks information about timbre. Both
formats include a large amount of information; machine learning tasks in MIR seek to reduce this
information and extract important features. A common approach to condensing the information
in audio formats is mel-frequency ceptstrum coefficiencs (MFCC). This paper presents a new alter-
native: a way of modeling MIDI information with a directed graph. This paper demonstrates that
eight features extracted from this graph alone are able to achieve .98 accuracy in a binary genre
classification task and .92 accuracy on a binary artist classification task. Furthermore, this paper
discusses potential alterations and improvements to the model as well as potential applications to
music generation.
A New Approach to Encryption Using Graph Isomorphism: [Repository]
I originally created Isomorph as an independent project for the Spring 2017 iteration of COMP15 at Tufts University, and I continued to develop it the following summer. With Isomorph, I demonstrate an approach to encryption that utilizes graph isomorphism. By representing files as graphs and leveraging the fact that each graph has N! isomorphic representations, this encryption scheme achieves a brute-force time complexity of O(N!). A formal paper has not yet been written, but a presentation of the project can be found on
Youtube.