We focus on the task of goal-oriented grasping, in which a robot is supposed to grasp pre-assigned goal object clutter and needs some pre-grasp actions such as pushes enable stable grasps. However, this task, gets positive rewards from environment only when successfully grasping object. Besides, joint pushing elongates action sequence, compounding problem reward delay. Thus, sample inefficiency...

We formulate and analyze a goal-oriented adaptive finite element method for semilinear elliptic PDE linear goal functional. The discretization is based on elements of arbitrary (but fixed) polynomial degree involves linearized dual problem. linearization part the proposed algorithm, which employs marking strategy different to that standard methods. Moreover, unlike well-known dual-weighted resi...