Berkeley AI Research · 2026-04-20模型评分49
GRASP:面向世界模型长时程规划的梯度规划器
Gradient-based Planning for World Models at Longer Horizons
导读
加州大学伯克利分校 人工智能(AI,Artificial Intelligence) 研究团队提出 GRASP,一种面向学习型世界模型(预测未来状态的模型)的梯度规划器,通过把轨迹提升到虚拟状态实现跨时间并行优化、在状态迭代中加入随机性做探索,并重塑梯度让动作获得清晰信号。该方法针对长时程规划中梯度爆炸或消失、非贪心路径陷入局部最优等问题,使基于梯度的规划更稳健。
分类置信度
高
- 官方或准官方来源,且无冲突类别。
- 正文、可信日期、主分类和证据摘录完整。
来源级别:官方一手
事件类型:研究发布
证据摘录
- “GRASP is a new gradient-based planner for learned dynamics (a “world model”) that makes long-horizon planning practical by (1) lifting the trajectory into virtual states so optimization is parallel across time, (2) adding stochasticity directly to the state iterates for exploration, and (3) reshaping gradients so actions get clean signals while we avoid brittle “state-input” gradients through high-dimensional vision models.”
来源:Berkeley AI Research · bair.berkeley.edu