Learning concepts with energy functions

In three linesOpenAI develops an energy-based model that learns spatial concepts (near, above, between, closest, furthest) from only five demonstrations. The model transfers concepts across domains: concepts learned in 2D particle environments applied to 3D physics-based robot tasks.

Summary generated by Claude — human-verified