Interested in machine learning and statistics for time series and graphs
Tromsø, Norway
Joined July 2019
- Tweets91
- Following148
- Followers298
- Likes251
How does the Research Council of Norway funding line up with publication and citation output across Norwegian institutions?
This dashboard lets you explore exactly that:
filippomb.github.io/NFR-fund…
Filippo Maria Bianchi retweeted
I will be in Vancouver next week for #ICML2025! If you’re around, would love to catch up 😊
Also, if you’re interested in uncertainty quantification in time series forecasting, come say hi at poster E-1706 on Tuesday from 16:30 to 19:00!
🚨 Exciting news! We released 🎱 tgp (Torch Geometric Pool), the library for pooling in Graph Neural Networks.
🚀 Get started with our tutorials: torch-geometric-pool.readthe…
With @IvanMarisca and Carlo Abate.
#GraphML #GNN #Pooling #Pyg
🚀 We just released the code for our new #ICML2025 paper "Relational Conformal Prediction for Correlated Time Series" by @andreacini1994 and co-authors!
💻 code: github.com/andreacini/corel
📑 paper: arxiv.org/abs/2502.09443
Filippo Maria Bianchi retweeted
Happy to share that CoRel has been accepted at ICML!🥳
CoRel quantifies uncertainty in correlated time series forecasting by leveraging graph deep learning operators. Check it out! 😎
W/ @alj_jenkins, Danilo, Cesare, and @FilippoMariaBi1
🔗 arxiv.org/abs/2502.09443
Tomorrow (April 26th), I'll present our recent work on MAXCUT with GNNs for graph pooling at #ICLR2025
Stop by to say hi at Poster 215 from 10:00 to 12:30 ☺️
Filippo Maria Bianchi retweeted
Join us to hear talks from great experts logml.ai/speaker.html
@marinkazitnik @tolga_birdal @FilippoMariaBi1 Coralia Cartis, Marco Cuturi, Estelle Massart, @Pseudomanifold, Francisco Ruiz, Alice Barbara Tumpach
Not happy with the review process at #ICML2025. 6 reviews in 1 month are too much: review quality and engagement in the rebuttal will be low.
Also, the 12 (!!) mandatory questions are often irrelevant. Why comment on the supplementary if a major flaw (e.g. plagiarism) is found?!
Filippo Maria Bianchi retweeted
Big personal update 😊
I'm moving to @UniofOxford to work with @mmbronstein on my recently funded Swiss National Science Foundation project 🎉🎉
As part of the project, I'll also be collaborating with @FilippoMariaBi1 😃
Looking forward to exciting new research & collabs! 😎
🎉 Our paper was accepted to #ICLR2025!
Project page: github.com/NGMLGroup/MaxCutP…
🔥 Bayesian nonparametrics meets pooling in GNNs! 🔥
This is the first clustering-based pooling method that learns and dynamically adapts the size of the pooled graph to input and downstream task.
📄 paper: arxiv.org/abs/2501.09821
💻 code: github.com/NGMLGroup/Bayesia…
Just published a new blog on 🎱 Pooling in #GraphNeuralNetworks!
💡Learn the fundamentals through this gentle introduction and discover how they can improve your GNN applications.
👉 Check it out here: gnn-pooling.notion.site/
#GNN #MachineLearning #AI
Filippo Maria Bianchi retweeted
Our new featured LoG2024 meetup will be held in the Artic❄️
Tromsø🇳🇴Meetup info:
Date-> 27 Nov. 2024
Website-> sites.google.com/view/log-tr…
Our new call for #LoG2024 local meetups is out! This "network" of local mini-conferences aims to bring together attendees belonging to the same geographic area, fostering discussions and collaborations.If you are interested in hosting one, please read this logconference.org/call-for-m…
Filippo Maria Bianchi retweeted
3. Graph Deep Learning for Time Series Processing: Forecasting, Reconstruction, and Analysis by @andreacini1994, @IvanMarisca, @dan_zambon
4. Integrating Knowledge Graphs and Large Language Models for Advancing Scientific Research by @qzhang_cs, @ChenJiaoyan1, @mengzaiqiao
🤔 How to interpret spatio-temporal data and deep learning models?
💡In our recent work with Michele Guerra and @s_scardapane we leverage Koopman theory to design an XAI framework for spatio-temporal GNNs.
📄 Preprint: arxiv.org/abs/2410.13469
💻 Code: github.com/NGMLGroup/Koopman…
Come to visit me and @IvanMarisca tomorrow at the first poster session (11:30-13:00 - Hall C) as we present our paper #1016!
#ICML2024
Filippo Maria Bianchi retweeted
🧵 Ready for #ICML2024!
This year me and @FilippoMariaBi1 present a method to forecast correlated time series with missing data. We compute a hierarchy of multi-scale spatiotemporal representations and adaptively combine them conditioned on the missing data pattern👇
🇫🇷 Next week I'll be in Paris attending #SIAMLA24.
I'll give a talk about our recent work on Total Variation Graph Neural Networks, first presented at ICML last year github.com/FilippoMB/Total-v…
Looking forward to discussions about graphs, machine learning, and optimization!
It was really a pleasure to be invited at the West Norway University of Applied Science as the opponent for the PhD thesis of Michele Gazzea.
Excellent work, both the defendant and his supervisor @RezaArghandeh!
Filippo Maria Bianchi retweeted
*Kolmogorov-Arnold Networks (KANs)* by @ZimingLiu11 et al.
Since everyone is talking about KANs, I wrote some notes on Notion with a few research questions I find interesting.
First time I do something like this, give me some feedback. 🙃
sscardapane.notion.site/Kolm…
