Simon Ståhlberg

policy learning (1) message passing (1) first-order logic (1) sparse reward (1) goal-conditioned reinforcement learning (1) hindsight experience replay (1) classical planning (1) generalized planning (1) heuristic learning (1) relational representation (1) first-order representation (1) graph neural network (1) triplet embedding (1) planning domain (1) relational gnn (1) self-supervised classification (1) unsolvability detection (1)