🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Mateo Espinosa's PhD explores how to improve the application of #machinelearning to #medicine - gatescambridge.org/about/news/mor… Cambridge Computer Science @Cambridge_uni St Edmund's College, Cambridge

Mateo Espinosa Zarlenga, Pietro Barbiero & Gabriele Ciravegna will describe concept-embedding models, a new self-explainable deep architecture that does not sacrifice accuracy for explainations (openreview.net/forum?id=HXCPA…) x.com/pietro_barbier…

Reunión de ecuatorianos en #NeurIPS2022 ! Con Ruben Villegas y Mateo Espinosa Zarlenga 🇪🇨🇪🇨🇪🇨

My new favorite thing - Bing's new ChatGPT bot argues with a user, gaslights them about the current year being 2022, says their phone might have a virus, and says "You have not been a good user" Why? Because the person asked where Avatar 2 is showing nearby

Concept-based systems, which map to human-interpretable concepts before final prediction, offer transparency + permit humans to intervene on preds. Practitioners working w/ such models often assume humans are certain oracles over concepts. Yet, real humans may be uncertain! 1/

“ Hypergraphs for computational genomics” makes the cover of Nature Machine Intelligence July issue! Congratulations Ramon, who’s an inspiring researcher and doesn’t like to boast on social media! nature.com/natmachintell/…

Wow, what a great group of mentors for the #LatinxinAI workshop at ICML Conference LatinX in AI (LXAI) #ICML2023

Creo que nunca ha habido tantos ecuatorianos reunidos en una conferencia como #ICML2023 Aquí estamos en el LatinX in AI LatinX in AI (LXAI) workshop, donde todos somos keynote speakers u organizadores! 🇪🇨💪🏾

That's a wrap on another great #LatinXinAI research workshop at ICML Conference thanks to the community who joined us! #ICML2023

🤖How can we account for human uncertainty when building new #AI models? Our researchers have developed a way to incorporate human uncertainty into systems where humans & AI work together, which will help boost trust & safety eng.cam.ac.uk/news/how-sure-… Katie Collins #ML

📢¡latin americans in AI!📢 pre-register before october 13th to attend a conference/summer-school in quito next february with some incredible speakers (pictured below). don't think you can afford it? we have travel grants, so apply now! riiaa.org/riiaa6

🚀 Thrilled to announce the KDD Workshop on Human-Interpretable AI! Join us to explore #AI explainability, aiming to surpass standard saliency maps and deliver explanations interpretable by all users. Submit your papers now and be part of shaping the future of #XAI! #KDD2024

🚨 Call for Workshop Papers - HI-AI at #KDD2024 🚨 🔍 Dive into the Workshop on Human-Interpretable AI! 🧠 Submit your work and join the efforts in constructing explanations aligned with human-understandable concepts! 📆 June 14 #XAI Gabriele Ciravegna human-interpretable-ai.github.io

🔥The Human-Interpretable AI workshop at #KDD2024 was a success! We had over 100 engaged researchers, 3 exceptional keynote speakers, and 15 insightful paper presentations. Thanks to all the co-chairs! 🎉 SIGKDD 2025

#neurips2024 will be very DEAR to me this year! Deep Equilibrium Algorithmic Reasoning has also been accepted at NeurIPS after its selection for an oral presentation at MAR@CVPR 2024. Special thanks to my collaborators Davide Buffelli Pietro Lio' and, my lucky charm, JJ Wilson!

Excited that our paper "Addressing Concept Mislabeling in Concept Bottleneck Models Through Preference Optimization" was accepted to ICML 2025! We show how Preference Optimization can reduce the impact of noisy concept labels in CBMs. 🧵/9

One of my favorite things at #ICML2025 was seeing so many Ecuadorians! These were taken at the LatinX in AI (LXAI) workshop at the start of the week. 🇪🇨