Talk today (Sunday 14th July) at oxfordml.school on causal effect estimation from observed data, with context and confounders! Slides here: gatsby.ucl.ac.uk/~gretton/paper…

👩‍🔬✨ Hear from our inspiring speakers, Clementine Domine and Ľudka Kučíková , as they will be delving into their groundbreaking work on deep learning theory, artificial neural networks and machine learning in neuroscience research. 🧵/2

It is going to be a busy week here in Vienna #ICML2024: - Wed 11.30 Poster 1704: curriculum vs overparameterisation; - Thu 11.30 Poster 1407: benefits of training with curricula in compositional tasks; - Sat @ GRaM workshop: bias from geometrical structure. More info below!

A distributional analogue to the successor representation! Tomorrow (Wednesday) at #ICML2024 during the afternoon poster session.

Distributional reinforcement learning when your return distribution is expressed as a mean embedding! Tomorrow (Wednesday) at #ICML2024 during the morning poster session.

Excited to present new work ICML Conference today at 13:30 with Yedi Zhang and Peter Latham on Understanding Unimodal Bias in Multimodal Deep Linear Networks. Find us at Hall C 4-9 #1502! 1/4

Given the energy spent improving predictive protein sequence models, can we find out how close they are to the ground truth? Log-likelihoods and RMSEs are no good for this, and can only be used for model comparison. (Deep) Kernel Methods to the rescue! (ICML 2024 Poster) 1/n

It’s today ICML Conference 25.07 Thursday 11.30-13.00 #1407 Come and chat! w/Stefano Sarao Mannelli - Hiring! Andrew Saxe

Come find out about unsupervised ground metric learning using tree 🌳 Wasserstein distance at GRaM Workshop at ICML 2024 ICML Conference tomorrow, work from my OIST internship with myamada0! 🌞

Check out our new paper, showcasing the power of equivariance to predict the binding between antigens and antibodies. 🧬🧫

Today is my first time #CCN2024. I'm very excited and mildly jetlagged! I'll be presenting two works: 📃B51 On compositional generalisation and curriculum learning; 📃C93 A framework for modelling critical learning. Looking forward to some feedback and follow-up ideas! 👇

📢Please share our dystonia survey! We would like to reach people living with Dystonia in the UK and sample their experiences of physiotherapy. Open for two months from today!

We started today the new edition of Analytical Connectionism! This week Linda Smith, Cengiz Pehlevan and Jonathan Cohen will give us in-depth lectures on topics ranging from development to analytical methods for NNs! Happy days! 🎈🎈🎈

🚀 Check out our latest work on oracle-free & convergent algorithms for learning *distributional* Successor Features! 📄 Paper: arxiv.org/abs/2409.00328 Great colab with Jesse Farebrother, Arthur Gretton, & Mark Rowland, from Mila - Institut québécois d'IA, McGill University, Google DeepMind, Gatsby Computational Neuroscience Unit.

Super excited to be apart of the first NeuroAI workshop Deep Learning Indaba in Dakar!! 🇸🇳 My favourite communities coming together 🤩 Congrats to the organising team Samuel Winbono-Mpaaba Mba, Abdel Mfougouon Njupoun, Cyril & Gertrude -- looking forward to the rest of the day and speakers Devon Jarvis & AZA Allsop, MD, PhD 😁