#artificial intelligence
Aug 2021
Subgoal Search For Complex Reasoning Tasks
It is shown that a simple approach of generating $k$-th step ahead subgoals is surprisingly efficient on three challenging domains: two popular puzzle games, Sokoban and the Rubik's Cube, and an inequality proving benchmark INT.
K. Czechowski, Tomasz Odrzyg'o'zd'z, Marek Zbysinski et al.
· Neural Information Processin... · 41 citations
· ⚡3