Helen Qu @_helenqu Twitter profile

Pinned Tweet

Helen Qu

6 months

I’m a Dr now!! so grateful to my advisor and all my collaborators, friends, and family for supporting me every step of the way 🥰

52

11

552

Last Seen Profiles

@SMFC_FEP

@ojj854

@MiniplayG70190

@sekai_card

@e_lukovnikov

$Rosa /ᐠ ܻ ⑅ܻ༵ ᐟ\$

@ditzysunflowerx

@ColtBoyKayy

@dizzydoesnteat

@Kuhn83130

@ALeague12541

@pipJXdhSRZ76BTm

@coleplante

@pen_house

@galery_basah10

@KnowledgeNolan

@JohnLeoNo1087

@AnotherMbd

@yo_onmall

@MeTheCookieLady

@AngelaHayn79984

@AmalaNovi

@mario96ct

@ttk401008224539

@mtc_57

@zhzh560583

@hirajayden

@KellyKincaidWWE

@Odysseum_store

@YadavJi____

@ogmaticio

@MlCI6RuZ96gMnbK

@MontiSagesBSB

@FnSabo1

@GrindHouseBB

@erzinsoysuz

@jandakembangstw

Helen Qu

@_helenqu

4 months

my thesis is now live on arXiv!

Helen Qu

@_helenqu

6 months

I’m a Dr now!! so grateful to my advisor and all my collaborators, friends, and family for supporting me every step of the way 🥰

52

11

552

8

18

343

Helen Qu

@_helenqu

1 year

How can choosing the wrong host galaxy for type Ia supernovae bias cosmology? 🌌 My new paper with the Dark Energy Survey collaboration investigates the impact of host galaxy mismatch on cosmology from 5 years of DES supernova data: , 🧵1/n

1

5

67

Helen Qu

@_helenqu

8 months

Excited to be joining @FlatironCCA @FlatironInst in the fall as a Flatiron Research Fellow! time to say goodbye to philly, my home for basically a decade (!!), and hello to nyc 🏙️

5

3

66

Helen Qu

@_helenqu

1 year

Can we infer redshift from supernova photometry alone? Introducing Photo-zSNthesis🌱, a deep learning-based approach for type Ia supernova redshift inference. We show >5x improvement on real data over existing SN photo-z approaches! , 🧵 1/5

1

8

64

Helen Qu

@_helenqu

1 year

The PLAsTiCC astronomical time series dataset is now on the Hugging Face Hub! 🤗 Now it only takes one line of code to download and integrate this dataset into your machine learning project - more details in the 🧵: 1/5

1

12

36

Helen Qu

@_helenqu

8 months

What's the best way to use unlabeled target data for unsupervised domain adaptation (UDA)? Introducing Connect Later: pretrain on unlabeled data + apply *targeted augmentations* designed for the dist shift during fine-tuning ➡️ SoTA UDA results! 🧵👇

1

5

23

Helen Qu

@_helenqu

7 months

today, gen AI performance is surprisingly robust to new data/tasks, even beating specialized models! the secret: training on large-scale unlabeled data. what can we as scientists learn from this? some thoughts on robustness & the power of the unlabeled data you already have:

1

11

Helen Qu

@_helenqu

1 year

I’ll be speaking at the @FlatironCCA @FlatironInst Cosmic Connections workshop on astrophysics x ML this week! Excited to learn about new directions in ML for cosmology, time domain science, and much more 💫

0

1

5

Helen Qu

@_helenqu

1 year

Check out our new overview of transformers/attention and a review of its applications in astronomy! Comments welcome 😊Thanks to @BhuvJain and Dimitrios for all their hard work!

Transformers for scientific data: a pedagogical review for astronomers

The deep learning architecture associated with ChatGPT and related generative AI products is known as transformers. Initially applied to Natural Language Processing, transformers and the...

arxiv.org

0

5

Helen Qu

@_helenqu

1 year

Finally, all credit goes to the original developers of PLAsTiCC incl @reneehlozek @lgalbany @emilleishida @gsnarayan @_sublunar_ and many others! I hope this is a step towards making ML in astro more accessible and reproducible. Feel free to reach out if you run into issues! 5/5

1

0

4

Helen Qu

@_helenqu

1 year

We show a >5x improvement on mean residuals over the widely used SN photo-z predictor, LCFIT+Z, on simulated/real SDSS + simulated LSST data! We also find that performance on real data is slightly diminished (compared to simulations) but still much better than the baseline. 4/5

1

0

4

Helen Qu

@_helenqu

7 months

Is pretraining all you need? We found that it sometimes fails to boost performance beyond no pretraining… We figured out why and developed Connect Later to boost accuracy and robustness with pretraining in all cases! 9/10

Connect Later: Improving Fine-tuning for Robustness with Targeted...

Models trained on a labeled source domain (e.g., labeled images from wildlife camera traps) often generalize poorly when deployed on an out-of-distribution (OOD) target domain (e.g., images from...

arxiv.org

1

3

Helen Qu

@_helenqu

1 year

Key results: - Δw = 0.0013 with pure SNIa sample (incl CMB prior) - 0.009 < Δw < 0.0032 with contamination from photometric classification This is <10% of expected total Δw for DES-SN5YR! However, we should keep an eye on this with shrinking stat+syst in future surveys. 4/n

1

0

3

Helen Qu

@_helenqu

1 year

Finally, so grateful to my advisor Masao Sako and amazing collaborators in the DES-SN working group for all the help I received along the way. Can’t believe >20 people read and commented on this paper 🙏 5/n

0

3

Helen Qu

@_helenqu

6 months

photo + afterparty credits to @astrojason20 🥰🥰

0

3

Helen Qu

@_helenqu

7 months

Connect Later gets state-of-the-art results on real-world astro classification, tumor detection, and wildlife identification tasks! From the WILDS robustness benchmark leaderboard (): 10/10

1

0

3

Helen Qu

@_helenqu

1 year

Key takeaways: - directional light radius method for host matching works - cuts on fitted lightcurve params decrease the mismatch rate (2.5% -> 1.7%) and severity (spread in Δz 1.1 -> 0.6) - redshift-dependent photometric classifiers are less accurate with wrong redshift 3/n

1

0

2

Helen Qu

@_helenqu

1 year

Our supernova redshifts come from their host galaxies, so incorrect host match -> wrong supernova redshift -> biased Hubble diagram. We designed DES5YR-like simulations to quantify the prevalence of host mismatches, their severity (Δz), and cosmology impact. 2/n

2

0

2

Helen Qu

@_helenqu

2 months

update: I'll be presenting Connect Later at #ICML2024 poster session 1! come chat with me at Hall C 4-9 poster #1012 !

5

0

2

Helen Qu

@_helenqu

1 year

@dr_guangtou thank you! the answer to the first question is what we were hoping to understand, as the presence of a trend would be more detrimental to cosmo estimates. spec follow-up would produce a redshift estimate but these spec zs are not as precise as galaxy spec z.

1

0

1

Helen Qu

@_helenqu

8 months

In the paper, we also describe a general framework for designing targeted augmentations. This design process provides an opportunity to naturally incorporate domain knowledge about the dataset! 4/5

1

0

1

Helen Qu

@_helenqu

1 year

Traditionally, SN redshifts come from host galaxies, but host galaxy redshifts can be unavailable, or host matching can be wrong (a teaser for another upcoming paper)! Photo-zSNthesis provides an alternative that's independent of host galaxy info. 2/5

1

0

1

Helen Qu

@_helenqu

1 year

Photo-zSNthesis uses images made from SN lightcurves, a format that makes redshift visible (see low/med/high redshift examples below)! These images are then processed by a convolutional neural network, which predicts a full redshift PDF for each SN. 3/5

1

0

1

Helen Qu

@_helenqu

1 year

Here's how to use it - only one (real) line of code! It uses the Hugging Face Datasets library (see for installation instructions) and works with both TensorFlow and PyTorch. Much easier than writing custom data loading/preprocessing scripts! 4/5

1

0

1

Helen Qu

@_helenqu

1 year

thanks to my advisor Masao Sako + acknowledged collaborators! 5/5

1

0

1

Helen Qu

@_helenqu

11 years

@spamrin @vanessawuhoo Mmm that's right.

1

0

1

Helen Qu

@_helenqu

1 year

First, some background on the PLAsTiCC (Photometric LSST Astronomical Time-series Classification Challenge) dataset: this is a set of ~3.5M simulated LSST lightcurves of 18 transient/variable sources including supernovae, AGN, etc. (see ) 2/5

1

0

1

Helen Qu

@_helenqu

11 years

@spamrin I'm so lucky he doesn't have a twitter.

0

1

Helen Qu

@_helenqu

1 year

Some use cases I can think of: developing classification/regression ML models (e.g. classification by object type or redshift prediction), projects for physics/data science courses, unsupervised learning (e.g. lightcurve modeling)... the possibilities are endless! 3/5

1

0

1