@apanisson

XAI, Fairness, Transparency: Doing research for Responsible AI, Machine Learning, Complex networks Principal Researcher at CENTAI. https://xitter.cf/t.co/hIEPZJV4qs

 Turin, Piedmont
andre.panisson.com
Joined June 2010
  • Tweets637
  • Following579
  • Followers849
42Media
🚨 This will be the last tweet from this account. 🚨 Future updates will now be posted only on Bluesky. This account will stay open only to prevent the handle from being taken by someone else. See you on the brighter side! 🌟 #BlueskySocial 🔵 panisson.bsky.social 🔵
110
André Panisson retweeted
Corrado Monti@c0rrad0m0nti15 Apr 2024
🔴 New paper on @TmlrPub! ML algorithms are everywhere – but often as black boxes obfuscating decisions. Hence explainers: you give them a model, they tell you what it's grounding its decisions on. But how to avoid explainers becoming black boxes? Who watches the watchmen? (1/3)
1
4
1
19
1,326
André Panisson retweeted
ECML PKDD@ECMLPKDD19 Jul 2023
Did you know #ECMLPKDD2023 will host the 1st Special Day on AI for Financial Crime Fight? Distinguished speakers will talk about fighting financial crimes with AI!
 The event is free, but with only limited seats! 📆 Sep 18 🔗2023.ecmlpkdd.org/ai4fcf 🎟️ai4fcf.eventbrite.it/
5
5
1,705
With our last pre-print showing a new fast and effective way for training GNNs, we are also releasing our "lightning-fast" implementation of uniform random spanning tree extraction: github.com/panisson/rst

GitHub - panisson/rst: RST is a Python package that leverages two fundamental algorithms for...

RST is a Python package that leverages two fundamental algorithms for generating Uniform Random Spanning Trees (RSTs). - panisson/rst

github.com
Francesco Bonchi@FrancescoBonchi13 Jun 2023
📢New pre-print alert🚨"Fast and Effective GNN Training with Linearized Random Spanning Trees" arxiv.org/pdf/2306.04828.pdf or how to avoid #Oversquashing and #Oversmoothing with @apanisson @FabioVi12599694 and @cla_gentile #GNN #MachineLearning
3
19
2,003
Oggi e domani CENTAI, il centro di ricerca avanzata nel campo #AI nato dalla partnership tra #IntesaSanpaolo ed un team internazionale di ricercatori, presenta Navigating the Research Frontier of AI and Complexity, dialoghi con la comunità scientifica. linkedin.com/feed/update/urn…

#community #stakeholders #goals #projects #ai #datascience #machinelearning #neuroscience #econom...

@CentaiInstitute is proud to announce Navigating the Research Frontier of AI and Complexity, its scientific kick-off event which will take place on March 1st and 2nd, at the Intesa Sanpaolo Skyscra...

linkedin.com
2
5
13
1,330
André Panisson retweeted
Yamir Moreno@cosnet_bifi1 Mar 2023
We just started @CentaiInstitute kick-off meeting with the first keynote lecture by Prof. J. Doyne Farmer.
2
5
30
3,107
Job opportunity! @CentaiInstitute is seeking talented individuals with background in #AI to join our team and tackle some pressing societal challenges. Help detect and combat coordinated inauthentic behavior while pursuing your #research goals. Apply now! centai.eu/news/call-for-posi…
5
12
1,365
Opening for a Postdoc position @CentaiInstitute to work on coordinated inauthentic behavior (e.g. misinformation, disinformation, propaganda). Do you enjoy academic research but would also like to have more real-world impact? Come work with us! centai.eu/news/call-for-posi… DM for info
16
23
2,585
📢Announcing CENTAI's scientific kick-off event next 1-2 of March, in Turin. Check-out the program and hurry up to register for free! Tickets are running out fast. centai.eu/news/navigating-th… #AI #ComplexSystems #Economics #MachineLearning #Neurosciences #DataScience
1
13
30
7,100
André Panisson retweeted
Michele Starnini@m_starnini22 Nov 2022
How does low-credibility information about vaccines spread across different countries on Twitter? In our last paper, we quantify cross-border misinformation flows between users exposed to anti-vaccination (no-vax) content, before and during COVID. 1/6 arxiv.org/pdf/2211.11495.pdf
2
8
1
22
New Research: Mapping urban socioeconomic inequalities in developing countries through Facebook advertising data frontiersin.org/articles/10.… #BigData
2
1
🚨We are hiring researchers 🚨 @CentaiInstitute at all levels of seniority (from post-docs to principal scientists) and in all areas of #AI #DataScience #MachineLearning and #ComplexSystems See centai.eu/careers/details or feel free to get in touch with me for more info #hiring
1
15
2
33
André Panisson retweeted
Corrado Monti@c0rrad0m0nti19 Oct 2022
🔴 Just presented a new paper about how users flow in political communities on Reddit @socinfo22 ! Our network shows which subreddits act as bridges toward liberal, leftist, or alt-right areas. Thanks to the team Cesare Rollo, @gdfm7 , @apanisson corradomonti.com/files/SocIn…
6
1
24
Check out this new title:Time-varying graph representation learning via higher-order skip-gram with negative sampling @EPJscience ift.tt/YcU43Gn

Time-varying graph representation learning via higher-order skip-gram with negative sampling

EPJ Data Science - Representation learning models for graphs are a successful family of techniques that project nodes into feature spaces that can be exploited by other machine learning algorithms....

link.springer.com
1
1
André Panisson retweeted
EPJ Data Science@epj_ds29 May 2022
Time-varying graph representation learning via higher-order skip-gram with negative sampling ift.tt/YcU43Gn

Time-varying graph representation learning via higher-order skip-gram with negative sampling

EPJ Data Science - Representation learning models for graphs are a successful family of techniques that project nodes into feature spaces that can be exploited by other machine learning algorithms....

link.springer.com
3
2
Our work "Time-varying Graph Representation Learning via Higher-Order Skip-Gram with Negative Sampling" with @simonepiaggesi has been published in @epj_ds! epjdatascience.springeropen.…

Time-varying graph representation learning via higher-order skip-gram with negative sampling

EPJ Data Science - Representation learning models for graphs are a successful family of techniques that project nodes into feature spaces that can be exploited by other machine learning algorithms....

link.springer.com
2
15
André Panisson retweeted
Yelena Mejova@yelenamejova5 May 2022
🔍 Did you Google something today? 🩺 Do you like healthcare? 👉 Read this: Googling for Abortion: How Google Mediates Access to Abortion Access -- and Why It Matters Now More Than Ever link.medium.com/fImV1jjZMpb #abortion #abortionrights #google #RoeVWade

Googling for Abortion: How Google Mediates Access to Abortion and Why It Matters Now More Than Ever

Study shows Google’s location search results reflect existing inequality in abortion access. What does this mean if Roe goes?

link.medium.com
2
12
1
19
🚨 Job openings in Copenhagen! 🚨3 PhD/postdoc positions in the @nerdsitu team at @ITUkbh on my COCOONS project on collective coordination. #ComputationalSocialScience #DataScience #NetworkScience #NLProc Details here: lajello.com/positions.html
Luca Maria Aiello (lajello.bsky.social)@lajello7 Dec 2021
I have been awarded the Carlsberg Young Researcher Fellowship! The project will research collective coordination in online social media using NLP and complex systems tools. Job calls coming soon! Thanks to @nerdsitu & @ITUkbh for the invaluable support.
2
50
8
79
André Panisson retweeted
paolo b@paolo_bajardi23 Feb 2022
New preprint out w @FrancescoBonchi @apanisson @notnotalan ! For graph classification tasks, why should your explanation language be forced to match the low-level input space of the black-box? Get a nice motif-based explanation with GraphShap! #xai arxiv.org/abs/2202.08815
1
5
16
André Panisson retweeted
Fondazione CRT@FondazioneCRT13 Jan 2022
Progetto Lagrange: aperto 𝐟𝐢𝐧𝐨 𝐚𝐥 𝟏𝟖 𝐟𝐞𝐛𝐛𝐫𝐚𝐢𝐨 il bando per 𝟖 𝐛𝐨𝐫𝐬𝐞 𝐝𝐢 𝐫𝐢𝐜𝐞𝐫𝐜𝐚 𝐚𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐚 in Data Science e Impatto Sociale. bit.ly/3tlC0Ng @ISI_Fondazione
4
4
Load more