First talk of the Causal Representation Learning workshop CRL workshop at NeurIPS 2023 at #NeurIPS2023 🎉@gemma_e_moran kicks of the first session with a talk on Identifiable representation learning via sparse decoding. #causality

I'll be giving an invited talk about "Nonparametric Causal Representation Learning from Multiple Environments" at the CRL Workshop CRL workshop at NeurIPS 2023 at 16:00 in Room 243-245.

🌟 Excited to welcome Julius von Kügelgen (Julius von Kügelgen) to OxML 2024! With a PhD in the Cambridge-Tübingen program and set to join ETH Zürich, their innovative research is breaking new ground in how we understand and apply causality in AI. Don't miss their insights into the power

It was nice to give a tutorial and hands-on session on "Causality for Machine Learning" at the INVICTA School this week. Thanks for having me! 🔗 Link to materials: github.com/INVICTA-School…

✨ ICLR2024 Spotlight ✨ Still want to identify causal variables in a partially observed setup? 🦥 We present a unified framework for studying the identifiability of representations learned from simultaneously, partially observed views, such as different data modalities. 🍍

To kick off the last day of MLx Health & Bio, Julius von Kügelgen (Julius von Kügelgen), from ETH Zürich (ETH Zürich), delivered a fascinating lecture on "Causal Representation Learning." 📝 As an associate researcher at the ETH AI Center (ETH AI Center), Julius provided deep insights into his

I'll be looking for 1-2 students this year (co-supervision included). See my website for details about my research and stay tuned for a page coming soon about my evaluation process. Apply to Mila by Dec 1!

ELLIS Institute Finland is launching with public (Opetus- ja kulttuuriministeriö) and private support (Peter Sarlin) 🤝 A significant boost for AI R&D and a signal that Finland is investing in AI to attract the best talents and fuel growth & expertise in Europe. fcai.fi/news/2024/10/2… ELLIS

✨New Preprint ✨ Ever thought that reconstructing masked pixels for image representation learning seems sub-optimal? In our new preprint, we show how masking principal components—rather than raw pixel patches— improves Masked Image Modelling (MIM). Find out more below 🧵

Call for Papers – CAR Workshop UAI 2025 in Rio de Janeiro, Brazil 🇧🇷 #uai2025 Are you pushing the boundaries of causal abstractions, causal representation learning, or clustered causal graphs? Submit to CAR 2025! Deadline: May 26, 2025 sites.google.com/view/car-25/

Quite excited about this work, led by the fantastic Jack Brady. If you're interested in theoretically-grounded {disentangled, compositional, object-centric} representation learning, check out the paper and come talk to us at the poster session :)