Woojoo Na

I am a 1st year PhD student at Northeastern University advised by Professor Jennifer Dy. My research interests are in the intersection of mathematics and machine learning. I’m interested in identifying and utilising algebraic and geometric properties in deep learning and optimization methods.

Previously, I received my B.A. in Mathematics at Oxford.

Before Northeastern, I have been advised by such wonderful advisors: Professor Thomas Lukasiewicz at Oxford-CS, Dr. Andrey Kormilitzin, at Oxford Mathematical Institute, Professor Christophe Petit at Oxford Mathematical Institute, and Professor Abiy Tasissa at Tufts University.

Email / CV / Google Scholar / GitHub

Timeline

12/2023: Graduated from Tufts University with MS in Computer Science.
07/2023: My paper, "RACH-Space: Reconstructing Adaptive Convex Hull Space with Applications in Weak Supervision" is now available
09/2021: Joined Tufts University as a Graduate Student
08/2021: Discharged from the Korean Army as Sergeant, after completing 18 months of service.
02/2020: Joined the Korean Military, to serve as FDC in the Artillery
01/2020: I presented our works on crypgraphy at CMS-Ewha Womens University
12/2019: We presented at 17th IMA International Conference on Cryptography and Coding
7/2019: Graduated from University of Oxford, BA(Hons) in Mathematics

Research

Deep learning & Optimization

Labeling unlabeled training data when only given partial, noisy information about the ground-truth label.

arXiv : "RACH-Space: Reconstructing Adaptive Convex Hull Space with Applications in Weak Supervision"
Weakly supervised learning

Developed a novel algorithm which labels data when given partial, noisy information about the ground-truth labels. State-of-the-art performance on real world benchmark data.

P2-equivariant Convolutional Networks on MNIST dataset

Machine Learning project, 2022
Geometric feature learning

We focused on the framework for CNNs that are equivariant under arbitrary group transformations, and looked at its applications and limitations, experimenting on the MNIST dataset.

Natural Language Processing

	Fake news detection and its impact on financial markets Brigade Commander Award, Republic of Korean Army Start-up competition (2021) Fake News Detection Devised an online platform for fake news detection using neural models, aimed at small investors prone to fake news. Won Brigade Commander Award for Excellence.
	Conceptual space of a transformer encoder Undergraduate Summer research Transformer encoder Worked on conceptual space of a transformer encoder, studying the similarities between NLP architectures and the human brain.
	2018 n2c2 challenge Woojoo Na, Andrey Kormilitzin Undergraduate Poster session 2018 n2c2 challenge In this challenge, we proposed a model for detecting Adverse Drug Events and Medication Extration in EHRs. We exploited CNN architecture and a character level word embedding model to demonstrate a robust approach to concept extraction and relation extraction.
	Knowledge graph completion Undergraduate Summer research Knowledge graphs Worked on knowledge graph completion for ontology reasoning.

Cryptography

Trapdoor attacks on Cayley hash function parameters
Alexander Allin, Woojoo Na, Christophe Petit
IMACC-17
IMACC 2019

By leveraging group properties in Cayley hash functions, we introduce a trapdoor attacks on Cayley hash function parameters.

Source code credit to Dr. Barron