Projects
Biological Plausibility in Modern Hopfield Networks
Jan 2020 - Dec 2022
This work was for my master's thesis; I investigated the biological plausibility of modern Hopfield Networks and developed a model of associative memory that is more biologically realistic.
Click here to read my thesis
Engagement and Trust: How People Process and Fact-Check Online News
May 2021 - Aug 2021
Designed a HCI/UX experiment that focused on how users navigate false information online. More specifically, to study whether there is a relationship between how engaging and trustworthy people perceive online content.
Click here for presentation
Leveraging Bayesian Neural Networks for Adversarial Attack and Defense
Jan 2021 - Apr 2021
This was a group project in which we investigated how Bayesian NNs could be used to defend Convolutional NNs (CNNs) against adversarial attacks. The CNNs were of the LeNet architecture, and the training and testing was done using the MNIST dataset.
Click here for GitHub repo
Numerical Solutions of the One-phase Classical Stephan Problem
Jan 2021 - Apr 2021
Used numerical methods in MatLab to model the transition of glaciers into water (i.e., the Stefan Problem). Analyzed solutions to this problem using three finite difference methods and compared the results to real data in the form of a paper and an oral presentation.
Click here for paper