TY - RPRT TI - Generation of Geodesics with Actor-Critic Reinforcement Learning to Predict Midpoints AU - Kazumi Kasaura PY - 2026 UR - https://arxiv.org/abs/2407.01991 ID - 2407.01991 ER -