Hi! I'm Roujia Wen,

a fourth-year student at Minerva Schools
studying physics and machine learning.
Intrigued by patterns & regularities in the world,
I'm eager to learn and explore.



physics

Self-organizing Behavior

collective motion of self-propelled particles (current)

I built a simulation software that integrated genetic algorithm with human-computer interactive discovery, using it to explore the mapping between physical parameters and collective patterns, and identify new behaviors.

web app   demo video
Quantum Information

numerical optimization in quantum information (summer 2018)

As part of the undergraduate research program at Perimeter Institute, I investigated the impact of formulations on numerical efficiency and stability for some linear programs in quantum information and provided an approach to certify solutions from numerical solvers.

talk video   slides
Muon Identification

deep learning in particle physics (fall semester 2017)

During my semester in Korea, I worked with the EPP group at Hanyang University and applied deep neural networks to particle detector data to improve the identification of isolated muons.

slides

computation

VAD Visual

scientific computing & visualizations (summer 2017)

I programmed a software which post-processes data from large scale biofluid dynamics simulations, calculates statistics and provides visualizations.

screenshots  poster
The SAT Problem

algorithm for NP-complete problem (summer 2016)

As an REU at Santa Fe Institute, I developed and optimized an algorithm for the boolean satisfiability problem (SAT) by applying novelty search and multi-objective optimization.

talk video   slides   paper
Cruise App

machine learning & recommender systems (gap year 2015)

I gathered tens of thousands of customer reviews, extracted ratings using machine learning methods and designed the recommendation algorithm for cruise matching.

web app

others

Wikilearn

Wikilearn

I worked with a small group of Minerva students on a project proposal in which we incorporate scientific learning tools into the content of Wikipedia. We finally presented our idea at the Wikimedia Headquarter in San Francisco.

talk video  slides