Data Science and Mining team @polytechnique. AI for Bigdata
LIX, École Polytechnique
Joined September 2013
- Tweets116
- Following34
- Followers133
- Likes21
Prof M. Vazirgiannis (@mvazirg) offers a keynote speech entitled "GNNs and Graph Generative models for ... - lix.polytechnique.fr/dascim/…
DaSciM team retweeted
Excited to share that our paper 'Weisfeiler and Leman go Hyperbolic: Learning Distance Preserving Node Representations' with @giannis_nikole and @mvazirg has been accepted at #AISTATS2023 @aistats_conf !
DaSciM team retweeted
Next Tuesday at noon, I will be presenting some of our work in the area of Graph Representation Learning. Please come join us if you are in the Saclay region.
Thank you very much @tomamoral and Victor-Emmanuel Brunel for the invitation and organisation!
📅 Register to the next #SeminairePalaisien on Tuesday 7th Feb at 12pm, with @JLutzeyer & Nabil H. Mustafa (@LipnLab)!
➡️ palaisien.fly.dev/register/2…
📍 @ENSAEparis
DaSciM team retweeted
#EMNLP2022
Excited to present my paper “Questioning the Validity of Summarization Datasets and Improving Their Factuality” in poster session 8 tomorrow 9:00-10:30 am. #NLProc
Many thanks to my advisors @mvazirg and @ChloeDClavel!
DaSciM team retweeted
Thrilled to announce that our paper "Graph Ordering Attention Networks" has been accepted at @RealAAAI #AAAI2023. Many thanks to my amazing co-authors @JLutzeyer , @GDasoulas and my advisor @mvazirg . (1/n)
🆙We are super excited to have @JLutzeyer continue in our team as Assistant Professor, after two years of brilliant postdoctoral research!!!
🎉A great evolution for DaSciM!!!
Starting September, I will be working as an Assistant Professor in @dascim_polytech @LIX_lab @Polytechnique @IP_Paris_!
I am excited by the prospect of working with graphs & GNNs for many years to come and to keep interacting with the brilliant @IP_Paris_ students & colleagues.
Congratulations to @giannis_nikole and @GDasoulas for their paper: “Permute Me Softly: Learning Soft Permutations for Graph Representations” accepted for publication to the prestigious TPAMI journal!!! Bravo!
Also congrats to the authors of previous papers accepted this month:
- "Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction" by @gsalhagalvan @JLutzeyer @GDasoulas accepted for publication in Elsevier's Neural Networks journal
- "Mass Enhanced Node Embeddings for Drug Repurposing" by @giannis_nikole , @MichailChatzia1 accepted in the ICML - Workshop on Computational Biology
Congratulations to all!
DaSciM team retweeted
New journal paper alert! 🚨
Congratulations to @gsalhagalvan, @JLutzeyer, @GDasoulas, R. Hennequin, and @mvazirg, for their paper on "Modularity-Aware Graph Autoencoders" accepted for publication in @ElsevierConnect's Neural Networks journal (IF: 8.05)!
@Deezer @Polytechnique
👉 Check out our group's work "Node Feature Kernels Increase Graph Convolutional Network Robustness" tomorrow at #AISTATS2022
Tomorrow 12:30-14:00 CET, @cmwu8, @melaseddik, @mvazirg and I will present our joint work at #AISTATS2022. Please join us there; we are excited to meet you!
arxiv.org/pdf/2109.01785.pdf
1/2
⬇️Check out this great talk by @JLutzeyer, postdoctoral researcher in our group!
Yesterday I had the pleasure of speaking at the workshop on "Recent Advances in Graph Machine Learning" (nkeriven.github.io/gml2022/). I would like to thank @n_keriven & E. Oyallon for the smooth organisation.
In case you are curious, here are my slides: bit.ly/3MBDlXn
[#Research] We are glad to co-organise the seminar series "Law, Society & AI", result of joint effort between @IP_Paris_ and @HECParis.
Talks take place online as well as on the @ParisSaclay campus. #AI #LegalTech
Check out the link here👉suchanek.name/work/research/…
DaSciM team retweeted
Great paper and thread!
- 😮that super simple MSE loss works vs. BEiT-style dVAE (multi-modal) cross-entropy
- <3 efficiency of asymmetric encoder/decoder
- 👏detailed training recipes
- +1 v curious about dataset size scaling
- bit of lack of commentary on test-time protocol
1/N The return of patch-based self-supervision! It never worked well and you had to bend over backwards with ResNets (I tried).
Now with ViT, very simple patch-based self-supervised pre-training rocks! First BeIT, now Masked AutoEncoders i1k=87.8% arxiv.org/pdf/2111.06377.pdf
🧶
[#EVENT] A huge thanks to Prof. #JohnIoannidis for his inspirational talk on "Meta-Research and the Quest for Better Science" @Polytechnique @IP__Paris 👏
A memorable afternoon full of enlightening insights and meaningful exchanges🎊🎆
#BREAKING We are super excited to host Prof. #JohnIoannidis, physician, writer and one of the top cited scientists in the world. He will give a talk entitled “Meta-Research and the Quest for Better Science” on November 10th @Polytechnique. Save the date!
👉bit.ly/2ZjO1WO
#EVENT Check out our demos and prototypes presented at “la journée « prototypes, logiciels et plateformes »” @IP__Paris.
(idia-pfday-2021.sciencesconf…)
GraKeL : ysig.github.io/GraKeL/0.1a8/
BARThez & BERTweetFR: master2-bigdata.polytechniqu…
DaSciM team retweeted
Pls check my keynote speech “Towards GNN explainability: Random Walk Graph Neural Networks” today at the @GReS_Workshop @ACMRecSys #RecSys2021 @IP_Paris @Polytechnique (europe.naverlabs.com/gres-wo…)
DaSciM team retweeted
Significant French language resources learned from large scale web/twitter crawl. Static/contextual word embeddings, N-grams, end2end demos (summarisation, title generation). Thanks to
@AgenceRecherche
chair HELAS funding master2-bigdata.polytechniqu……
@IP__Paris @Polytechnique