TY - RPRT TI - Learning to Manipulate Object Collections Using Grounded State Representations AU - Matthew Wilson AU - Tucker Hermans PY - 2020 UR - https://arxiv.org/abs/1909.07876 ID - 1909.07876 ER -