Zwee Dao

Startup | Product | AI

[rosie @therosieum]

https://t.co/F2G1usjT2E

[Zwee Dao @ZweeDao]

@elonmusk @Jihoz_Axie We at Cortex built a tool to visualize how your tweet spreads across Twitter. Interested?

[Ben Chamberlain @DrBPChamberlain]

1/4 Today we released two new link prediction models developed over the past 12m. We solve many problems that prevent GNNs succeeding at link prediction by adding a message passing mechanism that uses sketches of subgraphs.
https://t.co/0G6qTl5BiR

[Fabrizio Frasca @ffabffrasca]

Overwhelmed by a plethora of similar, expressive GNN architectures? Symmetries and equivariance to the rescue again! 🧵

A fresh, unifying perspective on Subgraph GNNs appearing at #NeurIPS2022 😊

Joint work with: @beabevi_ * @mmbronstein @HaggaiMaron – https://t.co/rpbEXXVqV1

[Emanuele Rossi @emaros96]

@Airbnb is using our SIGN method in production to compute graph-based user embeddings, leading to "notable performance gains".

Why was it chosen? Scalability, simplicity and predictive performance.

https://t.co/N1itGDSRqy
@AirbnbEng

[Zwee Dao @ZweeDao]

@play_eFootball Hi, could you please fix the button config on mobile. When I change the Defend buttons, the Attack buttons change too. Unplayable now. This did not happen in previous version.

[Zwee Dao @ZweeDao]

@emaros96 @elonmusk Without ML, hard for Twitter to fight spam or recommend “maximum fun” content

[Zwee Dao @ZweeDao]

@elonmusk Curious what kind of spam do you often experience on Twitter? I’m trying to solve that at Twitter.

[Zwee Dao @ZweeDao]

@elonmusk How about AI/ML? Can't imagine Twitter can detect spam or recommend maximum fun contents without AI/ML.

[Emanuele Rossi @emaros96]

The code for our Feature Propagation method to learn on graphs with partially missing node features is now available!
https://t.co/TBbkU08szM

Co-authors: @hennesseeeeee @migorinova @b_p_chamberlain @epomqo @mmbronstein @TwitterEng

[Michael Bronstein @mmbronstein]

New post @TDataScience about physics-inspired "continuous" graph ML models
🧵
https://t.co/A2EeSefUSI

[Emanuele Rossi @emaros96]

Excited to be presenting our work on GNNs for graphs with missing node features at the amazing Graph ML reading group hosted by @HannesStaerk!

See you tomorrow at 4pm GMT: https://t.co/z6M0QZ3MXO

Paper explanation: https://t.co/01PznbXnm7
Blog post: https://t.co/lSCNNQ9d0l

[Michael Bronstein @mmbronstein]

A recording of my talk @Harvard @HarvardCMSA showing our recent works @Twitter #Cortex on novel models for learning on graphs inspired by differential geometry, physics, and PDEs:

https://t.co/fCxj0jPxUg

[Emanuele Rossi @emaros96]

While most GNNs assume a full set of features for all nodes, real-world graphs often have partially missing node features. In a new blog post with @mmbronstein, we discuss how we can learn on such graphs using Feature Propagation (10 min read) 📝

https://t.co/lSCNNQ9d0l

[Michael Bronstein @mmbronstein]

An annual round of predictions in Geometric and Graph ML coauthored with @PetarV_93 based on input from the leading experts in the field.

Longread on @TDataScience

https://t.co/uSf4l6fRXx

[Claire @claireewoodcock]

I am beyond excited to present our research at @NeurIPSConf on Monday at 5pm EST. Come by to find out how real humans respond to AI explanations and how to make explanations better. Work done with @b_mittelstadt @danbusbridge and Dr Grant Blank.

[Zwee Dao @ZweeDao]

Created my first NFT on OpenSea 🤩 https://t.co/8GysXpPMBA via @opensea

[Engineering @Engineering]

Learn about the state of the art in Hyperparameter Tuning! @otterkoala from Twitter Cortex will present the follow up analysis of our 2020 Black Box Optimization Challenge at INFORMS on Oct 24:

Link below for more details about the event.
https://t.co/Z55ZfvJlrH