Twitter Beğeni Hilesi

Super excited to join EPFL as assistant professor in the fall of 2024. Heartfelt gratitude to everyone who supported me! I am #hiring #PhDs and #PostDocs! Interested in developing machine learning methods for enhancing science🧬and personalized medicine💊? Details in the🧵!

Equivariant neural nets are cool, but can be complicated and costly to implement. In our new NeurIPS Conference paper, we show how to transform your favorite trained neural net into an equivariant one 🔄It's cheap and easy! Come say hello at the poster session 👋🏾

I’ll be at #NeurIPS2023 presenting two of our workshop papers: 1. ICRM (arxiv.org/abs/2309.09888) at DistShift and R0-FoMo 2. CARE (arxiv.org/abs/2306.13924) at SSLTheoryPractice23 Please reach out if you’d like to chat about self-supervised learning and distribution shifts!

Excited to present a joint work with Stefanie Jegelka at NeurIPS 2023 Thursday morning poster session. Our paper (openreview.net/pdf?id=kDQwoss…) studies size transferability of graph neural networks with special attention to sparse (even bounded-degree) graph sequences. 1/4

Excited to present this work on Wednesday! Also, if you’re at NeurIPS and want to chat about equivariant learning, AI for science, or anything else, don’t hesitate to reach out 😄

2. "Sheaf-based Positional Encodings for Graph Neural Networks" (openreview.net/forum?id=ZtAab……) with Cristian Bodnar Pietro Lio': Another application of sheaves onto GNNs, where this time we construct positional encodings with the sheaf Laplacian😊.

Looking forward to mentoring a project at this great summer school! Students interested in GNNs / geometric deep learning should consider applying 😀

Excited to share three new papers accepted by #ICLR2024 for understanding self-supervised learning: 1. (Spotlight) On the Role of Discrete Tokenization in Visual Representation Learning 2. Do Generated Data Always Help Contrastive Learning? 3. Non-negative Contrastive Learning 🧵

SPE: On the Stability of Expressive Positional Encodings for Graph Neural Networks arxiv.org/abs/2310.02579

📢 PREPRINT ALERT! We study the relationship between enhanced expressivity and improved generalization performance in message-passing GNNs. This was a really fun collaboration with Billy Franks, Christopher Morris, and Floris Geerts. (1/3) arxiv.org/abs/2402.07568

Announcing Score Entropy Discrete Diffusion (SEDD) w/ Chenlin Meng Stefano Ermon. SEDD challenges the autoregressive language paradigm, beating GPT-2 on perplexity and quality! Arxiv: arxiv.org/abs/2310.16834 Code: github.com/louaaron/Score… Blog: aaronlou.com/blog/discrete-… 🧵1/n

Super excited to be releasing AlphaFold 3 today, developed by Isomorphic Labs and Google DeepMind: our next generation AI model for predicting the biomolecular structures and interactions of proteins, DNA, RNA, small molecules, and more: bit.ly/44yfaCw 1/

At #ICML I am presenting our position paper on end-to-end deep learning for relational databases! Stop by poster number #2302.

🚀 Announcing RelBench: an open benchmark for deep learning on relational databases! RelBench is the foundational infrastructure for research in Relational Deep Learning (RDL), which brings modern AI to structured data. RelBench has databases, tasks, loaders, evaluators, and

Less than a day to go to submit abstracts to Learning on Graphs Conference 2024 2024 -- send your work to LoG and we'll ensure you get expert reviewers who work in the same sub-field :) logconference.org/cfp/

Excited to announce @Entalpic has raised 8.5M€ in seed funding! 🎉 As CSO, this means taking my PhD research in AI-driven materials discovery out of the lab ⚗ and applying it to real-world industrial processes to reduce their carbon emissions🌍 Hiring: entalpic.notion.site/Entalpic-is-hi…