Very excited to be starting my PhD at MIT MIT Brain and Cognitive Sciences! I'll be working with Josh Tenenbaum on building bridges between cognitive science and AI.

Marco Baroni presents our #EMNLP2018 workshop paper on compositional learning, which is work João Loula did as an intern at FAIR

On International women's day, a reflection on gender bias in neuroscience and possible solutions with the amazing Ana Luísa Pinho

Come check out our work tomorrow at #CogSci2020

How can we learn representations for planning that are both efficient and flexible? Check out "Learning Constraint-Based Planning Models from Demonstrations" by @joaoloula, @kelseyrallen, and Tom Silver, on-demand for free at #IROS2020 iros2020.org/ondemand/

We are hiring a full time research assistant! This is a great opportunity to gain research experience at the intersection of cognitive science and AI, and hands-on experience managing a lab. Apply here: careers.peopleclick.com/careerscp/clie…

How do babies learn physical concepts like support and containment from just a few examples? We use simulation to bootstrap rule learning as a way of explaining what infants learn---and what they don't. Come watch our #cogsci2021 talk by @joaoloula and @kelseyrallen at 3pm at IP8

Our work, led by Pedro Tsividis, on theory-based reinforcement learning: arxiv.org/abs/2107.12544… CoCoSci MIT

Tackling complex problems with LMs requires search/planning, but how should test-time compute be structured? Introducing Self-Steering, a new meta-reasoning framework where LMs coordinate their own inference procedures by writing code!

Many LM applications may be formulated as targeting some (Boolean) constraint. Generate a… - Python program that passes a test suite - PDDL plan that satisfies a goal - CoT trajectory that yields a positive reward The list goes on… How can we efficiently satisfy these? 🧵👇

Our new work is out- Network Dynamics Theory of Human Intelligence biorxiv.org/content/early/…

Stochastic Subsampling for Factorizing Huge Matrices Our preprint with full proofs & empirical study (10x speedups) hal.archives-ouvertes.fr/hal-01431618v1…

My momma always said, "fMRI was like a box of chocolates. You never know what you're gonna get."

Great work João Loula ! x.com/uranus_2/statu…

Today we open-source Psychlab – a platform built on DeepMind Lab, designed to bridge the gap between AI and cognitive psychology. deepmind.com/blog/open-sour…

Work with Yaroslav Ganin , Igor, Ali Eslami and Oriol Vinyals. DeepRL + GANs + Program Synthesis + Robots x.com/OriolVinyalsML…

Announcing our first #AI research lab in continental Europe, #DeepMindParis, and a return to France for Remi Munos / Nous annonçons notre premier laboratoire de recherche #AI en Europe continentale, DeepMind Paris, et le retour de Rémi Munos en France: deepmind.com/blog/a-return-…