Ke Yu

Researcher in machine learning for healthcare. @UPitt

[Ricardo Martin-Brualla @rmbrualla]

My larger team at Google is hiring 2024 PhD interns working on generative media (video, image, 3D, etc)! Come work with great collaborators across Google.

See intern call below for details, links are available in the attached doc.

https://t.co/lUQ98kinAc

[Shantanu Ghosh @shantanuai]

#ICML2023 (1/n) Extremely happy to share our paper (Ke Yu, @ForoughArabsha1, @kbatmang ) has been accepted at ICML, 2023.

Project: https://t.co/5yoIAvsf8R
arXiv: https://t.co/wADp4rtYvN
code: https://t.co/pg3EXJTIio

[Kayhan Batmanghelich @kbatmang]

We are excited to host @martin_sen from @DeepMind in #MLxMed seminar tomorrow. He will discuss clinical challenges in moving #AI from code to clinic.
The talk is public; Plz RT
Wed 3-4 ET
Link https://t.co/eLIB6D5HJI
@GoogleHealth @Shyam_Vis @jeremycweiss @DBMI_Pitt #AI4Health

[Kayhan Batmanghelich @kbatmang]

On contrastive learning, shortcuts and some solutions to avoid it! It was fun working with these team @Josh_d_robinson @optiML @StefanieJegelka @KeonYu and Li Sun

[David Sontag @david_sontag]

How to beat linear models for ML on health data? Pre-train using them! See our #AAAI2021 paper on Reverse Distillation & a new transformer-based model (https://t.co/a9IASGM1bY). And... (drum roll) our new open-source code for ML on OMOP: https://t.co/gDg5z6JvXR @RBoiarsky @OHDSI

[Kayhan Batmanghelich @kbatmang]

The #MLxMed is back! Today’s seminar’s given by @ulasbagci about eye-tracking system to train a DL model for Computer Aided Diagnosis. Very excited about it! The talk is on Wed 3-4 ET. Public Zoom link: https://t.co/eLIB6D5HJI
@Shyam_Vis @jeremycweiss @DBMI_Pitt #MLforHealth

[JCIM & JCTC Journals @JCIM_JCTC]

#Semi_supervised Hierarchical Drug Embedding in #Hyperbolic Space
https://t.co/GgT2FAfiUx
@KeonYu @Shyam_Vis @DBMI_Pitt @Imaging_Gen
#current_issue #MachineLearning #ArtificialIntelligence

[Kayhan Batmanghelich @kbatmang]

The paper is accepted in @JCIM_ACS Journal.
We will discuss it in person Sat at #NeurIPS2020 workshop ML for molecules
code: https://t.co/lnriXUNwAX
Link: https://t.co/uLS5hPNpyj

@KeonYu @Shyam_Vis @DBMI_Pitt

[Kayhan Batmanghelich @kbatmang]

Indep. testing is one of the essential tools of #causal discovery. In reality, only a small subset of our samples have both X,Y while we have a much larger sample size unpaired data. @MingmingGong1 method (#NeurIPS2020 workshop) helps to be less data wasteful @DBMI_Pitt

[Ke Yu @KeonYu]

Our hyperbolic drug embedding paper is out on @JCIM_ACS: a method integrates drug taxonomy with chemical structure and enables localizing novel molecules in the context of the clinically approved drugs.
https://t.co/5H6Chn3tA8
@Imaging_Gen @Shyam_Vis

[Ke Yu @KeonYu]

Looking forward to the talk!

[Kayhan Batmanghelich @kbatmang]

We'll have another #MLxMed seminar by one of the pioneers of #ML4H @david_sontag. David discusses ML models that suggest treatment rather just Pro/Diagnosis. Excited about his talk on Wed 3-4 EST
Zoom link https://t.co/eLIB6D5HJI
@Shyam_Vis @jeremycweiss @DBMI_Pitt #MLforHealth

[arXiv Daily @Arxiv_Daily]

Semi-Supervised Hierarchical Drug Embedding in Hyperbolic Space
https://t.co/76mpaCZyFJ
by Ke Yu et al. including @Imaging_Gen
#ComputerScience #Learning