The Languini Kitchen serves as both a research collective and codebase
d...
Current state-of-the-art object-centric models use slots and attention-b...
Reinforcement learning agents must generalize beyond their training
expe...
How to improve generative modeling by better exploiting spatial regulari...
Common-sense physical reasoning in the real world requires learning abou...
Traditional sequential multi-object attention models rely on a recurrent...
First steps towards a mathematical theory of deep convolutional neural
n...