Action Jets：学习动作条件数据流形的局部形状
原标题：Action Jets: Learning the Local Shape of the Action-conditioned Data Manifold
AI 摘要
该论文提出 Action Jet 方法，通过估计动作条件数据流形的局部形状（泰勒多项式）来加速世界模型规划。固定状态后，不同动作对应的图像帧位于低维流形上，Action Jet 用几何编码器一次前向传播输出动作雅可比，并用校正头拟合二阶、三阶项。相比 IRASim 和 WorldGym 等生成器，对 16 个候选动作评分速度提升超过一个数量级，且在短时域预测更准确、计算量更少。
正文节选
Action Jets: Learning the Local Shape of the Action-conditioned Data Manifold Abstract World models are a powerful means of reasoning and planning, but many of these action-conditioned predictive models are high dimensional and computationally expensive. These challenges limit the number of action candidates that can be evaluated by a planner. We show that this cost is avoidable for planners over short horizons, which only require local information with respect to the current state. For a fixed