Modular meta-learning in abstract graph networks for combinatorial generalization

Understand

Modular meta-learning is a new framework that generalizes to unseen datasets by combining a small set of neural modules in different ways.

• In this work we propose abstract graph networks: using graphs as abstractions of a system's subparts without a fixed assignment of nodes to system subparts, for which we would need supervision.
• We combine this idea with modular meta-learning to get a flexible framework with combinatorial generalization to new tasks built in.
• We then use it to model the pushing of arbitrarily shaped objects from little or no training data.