Learning General Policies for Classical Planning Domains: Getting Beyond C$_2$

arXiv:2403.11734v1 Announce Type: cross
Abstract: GNN-based approaches for learning general policies across planning domains are limited by the expressive power of $C_2$, namely; first-order logic with two variables and counting. This limitation can be overcomed by transitioning to $k$-GNNs, for $k=3$, wherein object embeddings are substituted with triplet embeddings. Yet, while $3$-GNNs have the expressive power of $C_3$, unlike $1$- and $2$-GNNs that are confined to $C_2$, they require quartic time for message exchange and cubic space for embeddings, rendering …

abstract arxiv beyond cs.ai cs.lg domains embeddings general gnn gnns logic object planning power type variables