🙋♂️What is the wildest dream for graph foundation models?
🎯Graph across domains → a single model → all the downstream
🙋♀️Can we achieve that?
✅Yes! UniAug: Cross-Domain Graph Data Scaling with Diffusion Models
📃https://t.co/ofHT6Dus4C
Congratulations to CSE Ph.D. student Alex Rodriguez on his selection as a young researcher to the 9th Heidelberg Laureate Forum! Only 200 young researchers from around the world are selected to attend the forum in September!
https://t.co/Se4YG3IomS
Sharing another survey and reflective paper from 2021: I wrote a book chapter with @krvarshney, in which we ask the question "what are human-centered approaches doing for explainable AI"
https://t.co/Qaqt03I1wi
(a short 🧵)
Looking for material to dive into geometric deep learning?
As part of the @AIMS_Next #AMMI2021, we (@mmbronstein@joanbruna@TacoCohen) have delivered a course on GDL, which closely follows our proto-book. We make all materials publicly available!
https://t.co/ZsBDVa82Cp
Today we’re joined by @yuqirose, assistant professor at the @UCSDJacobs at @UCSanDiego to discuss her research on advancing ML algorithms for analyzing large scale time-series and spatiotemporal data.
🎧: Chartable Link https://t.co/OKmzz3VLKR
🎥: YT Link https://t.co/JSoZqTHH8o
The workplace hellscape of algorithmic injustice: punitive automated decisions that ignore context, lack of due process, predictable facial recognition fails, undue constant anxiety, and corporate apathy to mistreated workers perceived as expendable.
https://t.co/lgrGYvM4o6
New preprint with my colleagues @jpmorgan AI Research including @tl_spooner!
We use Bayesian optimization to generate counterfactual explanations that work for both regression and classification models.
Excited for Networked Justice symposium tomorrow morning where I'll be presenting on "Graph Augmentation for Equitable Access to Networked Resources" https://t.co/48Ywt9BIYe [joint work with @sachin_logs@ivanbrugere@CaimingXiong@SFResearch] @CSL_Illinois
4K version of my ICLR keynote on #geometricdeeplearning is now on YouTube:
https://t.co/evlEnDlX8E
Accompanying paper: https://t.co/LgeCCZMaqE
Blog post: https://t.co/U77kwqqt4t
Excited to announce that our tutorial with @darbour26 on "Causal inference from network data" has been accepted as a lecture-style tutorial at #KDD2021. Stay tuned for more details: https://t.co/BqGsbLZn6p. @kdd_news
You can now find all the papers, code, data, slides and talk videos on our work on measuring and giving feedback on expressed empathy here:
https://t.co/2UmZfRHuLX
Big thanks to @TalkLifeApp and @MSFTResearch that made all this possible! @TheWebConf#WWW2021
At #ICLR2021 we will present the *updated* “BERTology🔤meets Biology🧬: Interpreting Attention in Protein Language Models": https://t.co/ZrHs8UoaBr
Trained only w/ language modeling, the Transformer's attention mechanism recovers folding structure & other protein properties. 1/N
This week I will be speaking:
- Tuesday 4/26 4pm PDT at @UCLA CS 201 on Fairness in Financial Services (thanks to @QuanquanGu) https://t.co/AVq3mc0Wg2
- Thursday 4/28 5.30 pm EDT at @JPMorgan Tech for Social Good LGBTQ+ recruiting https://t.co/AQtiq3BNwQ
Watch live at https://t.co/zyyhvYvizO #TheWebConf 2021 evening keynotes for Tuesday, April 21 are Yejin Choi from University of Washington and Estevam Hruschka from Megagon Labs #WWW21@marko_grobelnik