• Corpus ID: 8140719

From Non-Negative to General Operator Cost Partitioning

@inproceedings{Pommerening2015FromNT,
  title={From Non-Negative to General Operator Cost Partitioning},
  author={F. Pommerening and Malte Helmert and Gabriele R{\"o}ger and Jendrik Seipp},
  booktitle={AAAI},
  year={2015}
}
Operator cost partitioning is a well-known technique to make admissible heuristics additive by distributing the operator costs among individual heuristics. [] Key Method We show that LP heuristics for operator-counting constraints are cost-partitioned heuristics and that the state equation heuristic computes a cost partitioning over atomic projections. We also introduce a new family of potential heuristics and show their relationship to general cost partitioning.

Figures and Tables from this paper

A Comparison of Cost Partitioning Algorithms for Optimal Classical Planning
TLDR
This work evaluates these algorithms experimentally on pattern databases, Cartesian abstractions and landmark heuristics, showing that saturated cost partitioning is usually the method of choice on the IPC benchmark suite.
Lagrangian Decomposition for Optimal Cost Partitioning
TLDR
An anytime algorithm is derived to compute an optimal non-negative cost partitioning of abstraction heuristics without involving an LP solver and the extension to general cost functions is discussed.
New perspectives on cost partitioning for optimal classical planning
TLDR
It is proved that non-negative cost functions are not necessary, and that allowing negative values as well makes the framework more powerful: the resulting heuristic values can be exponentially higher, and unsolvability can be detected even if all component heuristics have a finite value.
Narrowing the Gap Between Saturated and Optimal Cost Partitioning for Classical Planning
TLDR
It is shown that searching in the space of orders leading to significantly better heuristic estimates than with previously considered orders and using multiple orders leads to a heuristic that is significantly better informed than any single-order heuristic.
Subset-Saturated Transition Cost Partitioning for Optimal Classical Planning
Cost Partitionings allow additively combining the heuristic estimates of multiple abstractions. This often gives better heuristic estimates compared to a single abstraction heuristic.
Subset-Saturated Transition Cost Partitioning
TLDR
This work shows that the computational overhead caused by allowing more expressive cost assignments is often too large for the additional heuristic accuracy to lead to solving more tasks and generalized saturated operator cost partitioning in a different direction.
Subset-Saturated Cost Partitioning for Optimal Classical Planning
TLDR
This work generalizes saturated cost partitioning by allowing to preserve the heuristic values of only a subset of states and shows that this often leads to stronger heuristics.
Lagrangian Decomposition for Classical Planning (Extended Abstract)
TLDR
This work analyzes the application of Lagrangian decomposition, a classical tool in mathematical programming, to cost partitioning of operator-counting heuristics to view the computation as an iterative process that can be seeded with any cost partitions and that improves over time.
Saturated Cost Partitioning for Optimal Classical Planning
TLDR
This work proposes a greedy algorithm to generate orders and shows how to use hill-climbing search to optimize a given order, which leads to significantly better heuristic estimates than using the best random order that is generated in the same time.
Lagrangian Decomposition for Classical Planning
TLDR
This work analyzes the application of Lagrangian decomposition, a classical tool in mathematical programming, to cost partitioning of operator-counting heuristics to view the computation as an iterative process that can be seeded with any cost partitions and that improves over time.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 23 REFERENCES
LP-Based Heuristics for Cost-Optimal Planning
TLDR
With this new method of analysis, dominance of the recent LP-based state-equation heuristic over optimal cost partitioning on single-variable abstractions is shown and it is shown that the previously suggested extension of the state- EQUATION heuristic to exploit safe variables cannot lead to an improved heuristic estimate.
Optimal admissible composition of abstraction heuristics
New Admissible Heuristics for Domain-Independent Planning
TLDR
Methods for automatically deriving additive hm and PDB heuristics from STRIPS encodings are advanced and improvement over existing heuristic in several domains is shown, although, not surprisingly, no heuristic dominates all the others over all domains.
An Admissible Heuristic for SAS+ Planning Obtained from the State Equation
TLDR
A new admissible heuristic is obtained from the state equation associated to the Petri-net representation of the planning problem that can be computed in polynomial time and is competitive with the current state of the art for optimal planning, as empirically demonstrated over a large number of problems.
Cost-Optimal Planning with Landmarks
TLDR
This work proposes a methodology for deriving admissible heuristic estimates for cost-optimal planning from a set of planning landmarks, and presents a simple best-first search procedure exploiting such heuristics.
Flow-Based Heuristics for Optimal Planning: Landmarks and Merges
TLDR
A flow-based heuristic for optimal planning that exploits landmarks and merges that allows us to partially and incrementally merge variables, thus avoiding the formation of cross products of domains as done when merging variables in the traditional way.
The Fast Downward Planning System
  • M. Helmert
  • Computer Science
    J. Artif. Intell. Res.
  • 2006
TLDR
A full account of Fast Downward's approach to solving multivalued planning tasks is given and a new non-heuristic search algorithm called focused iterative-broadening search, which utilizes the information encoded in causal graphs in a novel way is presented.
A Formal Basis for the Heuristic Determination of Minimum Cost Paths
TLDR
How heuristic information from the problem domain can be incorporated into a formal mathematical theory of graph searching is described and an optimality property of a class of search strategies is demonstrated.
Getting the Most Out of Pattern Databases for Classical Planning
TLDR
It is shown how stronger heuristic estimates can be obtained through linear programming, combining information from multiple patterns in the so-called canonical heuristic.
Disjoint pattern database heuristics
...
1
2
3
...