Abstract
Polynomial-time heuristic functions for planning
are commonplace since 20 years. But polynomialtime in which input? Almost all existing approaches are based on a grounded task representation, not on the actual PDDL input which is exponentially smaller. This limits practical applicability to cases where the grounded representation
is “small enough”. Previous attempts to tackle this
problem for the delete relaxation leveraged symmetries to reduce the blow-up. Here we take a more
radical approach, applying an additional relaxation
to obtain a heuristic function that runs in time polynomial in the size of the PDDL input. Our relaxation splits the predicates into smaller predicates
of fixed arity K. We show that computing a relaxed plan is still NP-hard (in PDDL input size) for
K ≥ 2, but is polynomial-time for K = 1. We implement a heuristic function for K = 1 and show
that it can improve the state of the art on benchmarks whose grounded representation is large.
are commonplace since 20 years. But polynomialtime in which input? Almost all existing approaches are based on a grounded task representation, not on the actual PDDL input which is exponentially smaller. This limits practical applicability to cases where the grounded representation
is “small enough”. Previous attempts to tackle this
problem for the delete relaxation leveraged symmetries to reduce the blow-up. Here we take a more
radical approach, applying an additional relaxation
to obtain a heuristic function that runs in time polynomial in the size of the PDDL input. Our relaxation splits the predicates into smaller predicates
of fixed arity K. We show that computing a relaxed plan is still NP-hard (in PDDL input size) for
K ≥ 2, but is polynomial-time for K = 1. We implement a heuristic function for K = 1 and show
that it can improve the state of the art on benchmarks whose grounded representation is large.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. IJCAI-21 |
| Publisher | International Joint Conferences on Artificial Intelligence |
| Publication date | Jul 2021 |
| Pages | 4119-4126 |
| ISBN (Electronic) | 978-0-9992411-9-6 |
| Publication status | Published - Jul 2021 |
| Event | Thirtieth International Joint Conference on Artificial Intelligence. IJCAI-21 - Montreal-theme Virtual Reality, Canada Duration: 19 Aug 2021 → 26 Aug 2021 |
Conference
| Conference | Thirtieth International Joint Conference on Artificial Intelligence. IJCAI-21 |
|---|---|
| Location | Montreal-theme Virtual Reality |
| Country/Territory | Canada |
| Period | 19/08/2021 → 26/08/2021 |