Hi, I'm Usha! I'm an incoming first-year PhD student at Harvard, coadvised by Prof. Hima Lakkaraju and Prof. Hanspeter Pfister. I go by she/her and they/them pronouns.

Get to know me better!

- email: usha[_]bhalla[@]g[.]harvard[.]edu

- cv: [link]

image of me


My research interests generally lie in the realm of machine learning/computer vision and 1) their applications to healthcare and sustainability, 2) improving the interpretability of models, and 3) improving robustness to limited, biased, or unreliable datasets.
This is a collection of some of my work and research experiences thus far.

* Microsoft Research Intern

Developed a 4D-spatiotemporal neural network robust to domain shifts and limited data to identify which scenes of a movie people will recall later on using fMRI scans taken during watch time.

* Brown Honors Thesis

Currently pursuing a Senior Thesis with Professor Chen Sun on the topic of increasing model interpretability with limited supervision.

[T. Yun, U. Bhalla, E. Pavlick, C. Sun. “Do Vision-Language Pretrained Models Learn Primitive Concepts?” arXiv.]

* Rosetta Commons REU Intern

Applied Procedural Content Generation via Machine Learning to the tutorial of Foldit, a game about protein folding and design used to teach biochemstry, to create computer generate levels tailored for a specific user's difficulty, as a part of the NSF-funded Rosetta Commons Research Experience.

[U. Bhalla, A. Sarkar, S. Cooper. “Application of Procedural Content Generation via Machine Learning to the Tutorial Levels of Foldit.” RosettaCon 2020.]

* Barrow Neurological Institute Translation Bioimaging Intern

Analyzed the accuracy and precision of recommended protocols for the acquisition of DSC-MRI images of glioblastomas, metastases, and lymphomas as an intern for Barrow’s Research Program

[N. Semmineh*, U. Bhalla*, C. Quarles, et al. “Analysis of Accuracy and Precision of Recommended Protocols for Dynamic Susceptibility Contrast MRI for Brain Metastases.” ISMRM 2020.]

* ASU SGSUP Summer Intern

Analyzed the correlation between the urban heat effect and heat related deaths in Maricopa County with data from the Maricopa County Department of Public Health, National Centers for Environmental Information, Maricopa County Air Quality Department, and the Arizona Meteorological Network

[David M. Hondula, Aleš Urban, and Usha Bhalla. "A Novel Method for Assessing the Health Impacts of Urban Heat". UWIN 2017.]


* Projects

- AD Finder

My final project for CS 1470 - Deep Learning @ Brown. Built an RNN-based neural network to aid in Alzheimer's classification using speech transcript data.

[Code] [Paper]

- Recolorer

My final project for CS 1430 - Computer Vision @ Brown. Built an app that smartly recolors artwork and imagesbased on user input palettes

[Code] [Paper]

- Sea Shell

My final project for CS 33 - Computer Systems @ Brown. Built a command terminal shell entirely in C

Code: available on request

* Teaching

- Head Teaching Assistant

Develop course material, hire + head course staff, and fulfill UTA duties

- Undergraduate Teaching Assistant

Grade student work, hold TA hours, lead recitation sections, create and maintain course website

* Web Design

- Meal Mantra

Designed and launched current website for Meal Mantra.

- ISS + Embolden

Independently designed and developed website and marketing materials for Integrated Staffing Solutions, leading to an increase in applicants of over 1000%.


* Illustration

My portfolio of sorts is here!

* Dance

- Brown Abhinaya

I am a captain and choreographer for Brown's premiere classical south asian fusion group!

Look at some of our performances here, and follow us on instagram!

- Arangetram

Under the tutelage of Smt. Shyla Ganesan of Nrithyalaya Aesthetics Society, my sister and I completed our arangetram December 2017. You can watch my favorite piece, a varnam on dashavatar, here!

* Brown Daily Herald

I have been the Graphics Editor for the BDH since September 2019. The majority of my work is here!