Helen Qu

helenqu [at] sas.upenn.edu

I am an incoming research fellow at the Flatiron Institute. Previously, I completed my PhD in 2024 at the University of Pennsylvania, advised by Dr. Masao Sako. I work on robust machine learning and its applications to cosmology, specifically with type Ia supernovae as cosmological probes. I am part of the Dark Energy Survey Collaboration as well as the Nancy Grace Roman Space Telescope supernova science investigation team. I am particularly interested in developing deep learning methods with large-scale astronomical survey data applications in mind.

Selected Publications

Connect Later: Improving Fine-tuning for Robustness with Targeted Augmentations
NeurIPS 2023 Workshop on Distribution Shifts
Helen Qu, Sang Michael Xie

The Dark Energy Survey Supernova Program: Cosmological Biases from Host Galaxy Mismatch of Type Ia Supernovae
Astrophysical Journal, 2024
Helen Qu, Masao Sako, Maria Vincenzi, et al. + DES Collaboration

Photo-zSNthesis: Converting Type Ia Supernova Lightcurves to Redshift Estimates via Deep Learning
Astrophysical Journal, 2023
Helen Qu, Masao Sako

A Convolutional Neural Network Approach to Supernova Time-Series Classification
ICML 2022 Workshop on Machine Learning for Astrophysics
Helen Qu, Masao Sako, Anais Möller, Cyrille Doux

Photometric Classification of Early-Time Supernova Lightcurves with SCONE
Astronomical Journal, 2022
Helen Qu, Masao Sako

SCONE: Supernova Classification with a Convolutional Neural Network
Astronomical Journal, 2021
Helen Qu, Masao Sako, Anais Möller, Cyrille Doux

Sum-of-Parts Models: Faithful Attributions for Groups of Features
NeurIPS 2023 Workshop on Explainable AI in Action (XAIA)
Weiqiu You, Helen Qu, Marco Gatti, Bhuvnesh Jain, Eric Wong

The Dark Energy Survey: Cosmology Results With~ 1500 New High-redshift Type Ia Supernovae Using The Full 5-year Dataset
Astrophysical Journal, 2024
The Dark Energy Survey Collaboration (incl. Helen Qu)

Transformers for scientific data: a pedagogical review for astronomers
arXiv, 2023
Dimitrios Tanoglidis, Bhuvnesh Jain, Helen Qu

What’s the Difference? The Potential for Convolutional Neural Networks for Transient Detection without Template Subtraction
Astronomical Journal, 2023
Tatiana Acero-Cuellar, Federica Bianco, Greg Dobler, Masao Sako, Helen Qu

The Pantheon+ Analysis: Cosmological Constraints
Astrophysical Journal, 2022
Dillon Brout, Dan Scolnic, Brodie Popovic, …, Helen Qu, et al.

Selected Talks

Enabling Time Domain Science: From CNNs to Foundation Models
Flatiron Institute ML x Astrophysics Symposium (May 2023)

A Convolutional Neural Network Approach to Supernova Time-Series Classification
Rising Stars in Data Science (November 2022)

A Convolutional Neural Network Approach to Supernova Time-Series Classification
LSST Bayesian Deep Learning Workshop (June 2022)

Building the Toolbox for Type Ia Supernova Cosmology
Invited Seminar Talk, University of Delaware (March 2022)

Photometric Supernova Classification in the Roman Era
Roman Time Domain Conference (February 2022)

SN Ia Cosmology and Core-Collapse Science with the Roman Space Telescope
American Astronomical Society meeting (January 2022)

Misc.

I completed a BSE in Computer Science at Penn. I have also spent some time as a software engineer at Academia.edu, Internet.org by Facebook, and Yahoo.

Outside of research, I love to bake and eat and occasionally make music.