Fetching the paper…

DIRL: Domain-Invariant Representation Learning for Sim-to-Real Transfer · Around