Pattern Databases for Stochastic Shortest Path Problems

Klößner, Thorsten; Hoffmann, Jörg

doi:10.1609/socs.v12i1.18561

Cited by 3 publications

(9 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…h n (s)]}. Moreover, informative domain-independent heuristics for single objective SSPs are also relatively new (Trevizan, Thiébaux, and Haslum 2017;Klößner and Hoffmann 2021): a common practice was to resort to classical planning heuristics obtained after determinisation of the SSP (Jimenez, Coles, and Smith 2006).…”

Section: Heuristicsmentioning

confidence: 99%

“…• A more elaborate option is to consider only the stochastic aspects of the problem, resulting in an ideal-point SSP heuristic. In our experiments, we apply the recent SSP canonical PDB abstraction heuristic by Klößner and Hoffmann (2021) to each objective which we call H pdb2 ideal and H pdb3 ideal for patterns of size 2 and 3, respectively. • Alternatively, one might consider only the multiobjective aspects, by applying some of the MO deterministic heuristics (Geißer et al 2022) to the determinised SSP.…”

Section: Heuristicsmentioning

confidence: 99%

“…• The best option is to combine the power of SSP and MO heuristics. We do so with a novel heuristic H pdb mossp which extends the SSP PDB abstraction heuristic (Klößner and Hoffmann 2021) to the MO case by using an MO extension of topological VI (TVI) (Dai et al 2014) to solve each projection and the comax operator to combine the results. Our experiments use H pdb2 mossp and H pdb3 mossp .…”

Section: Heuristicsmentioning

confidence: 99%

“…We also consider the problem of guiding the search of these algorithms with domain-independent heuristics. A plethora of domain-independent heuristics exist for classical planning, but works on constructing heuristics for (single objective) probabilistic or multi-objective (deterministic) planning are much more recent (Trevizan, Thiébaux, and Haslum 2017;Klößner and Hoffmann 2021;Geißer et al 2022). Building on these recent works, we investigate a spectrum of heuristics for MOSSPs differing in their ability to account for the probabilistic and multi-objective features of the problem.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Heuristic Search for Multi-Objective Probabilistic Planning

Chen

Trevizan

Thiébaux

2023

AAAI

View full text Add to dashboard Cite

Heuristic search is a powerful approach that has successfully been applied to a broad class of planning problems, including classical planning, multi-objective planning, and probabilistic planning modelled as a stochastic shortest path (SSP) problem. Here, we extend the reach of heuristic search to a more expressive class of problems, namely multi-objective stochastic shortest paths (MOSSPs), which require computing a coverage set of non-dominated policies. We design new heuristic search algorithms MOLAO* and MOLRTDP, which extend well-known SSP algorithms to the multi-objective case. We further construct a spectrum of domain-independent heuristic functions differing in their ability to take into account the stochastic and multi-objective features of the problem to guide the search. Our experiments demonstrate the benefits of these algorithms and the relative merits of the heuristics.

show abstract

Section: Heuristicsmentioning

confidence: 99%

Section: Heuristicsmentioning

confidence: 99%

Section: Heuristicsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Heuristic Search for Multi-Objective Probabilistic Planning

Chen

Trevizan

Thiébaux

2023

AAAI

View full text Add to dashboard Cite

show abstract

“…Heuristic search algorithms (Hansen and Zilberstein 2001;Bonet and Geffner 2003; can solve SSPs without constructing the full state space by utilizing an admissible heuristic, i. e., a function that underestimates the expected cost-to-goal of all states. Previous work on admissible heuristics include determinizationbased approaches (e. g., Bonet and Geffner 2005;Weld 2010, 2012), which evaluate classical planning heuristics on the all-outcomes determinization of the problem; occupation measure heuristics (Trevizan, Thiébaux, and Haslum 2017), which can be seen as a generalization of operator-counting heuristics from classical planning (Pommerening et al 2014); and pattern database heuristics for SSPs (Klößner and Hoffmann 2021).…”

Section: Introductionmentioning

confidence: 99%

A Theory of Merge-and-Shrink for Stochastic Shortest Path Problems

Klößner

Torralba

Steinmetz

et al. 2023

ICAPS

View full text Add to dashboard Cite

The merge-and-shrink framework is a powerful tool to construct state space abstractions based on factored representations. One of its core applications in classical planning is the construction of admissible abstraction heuristics. In this paper, we develop a compositional theory of merge-and-shrink in the context of probabilistic planning, focusing on stochastic shortest path problems (SSPs). As the basis for this development, we contribute a novel factored state space model for SSPs. We show how general transformations, including abstractions, can be formulated on this model to derive admissible and/or perfect heuristics. To formalize the merge-and-shrink framework for SSPs, we transfer the fundamental merge-and-shrink transformations from the classical setting: shrinking, merging, and label reduction. We analyze the formal properties of these transformations in detail and show how the conditions under which shrinking and label reduction lead to perfect heuristics can be extended to the SSP setting.

show abstract

Cartesian Abstractions and Saturated Cost Partitioning in Probabilistic Planning

Klößner,

Seipp,

Steinmetz

2023

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

Stochastic shortest path problems (SSPs) capture probabilistic planning tasks with the objective of minimizing expected cost until reaching the goal. One of the strongest methods to solve SSPs optimally is heuristic search guided by an admissible (lower-bounding) heuristic function. Recently, probability-aware pattern database (PDB) abstractions have been highlighted as an efficient way of generating such lower bounds, with significant advantages over traditional determinization-based approaches. Here, we follow this work, yet consider a more general type, Cartesian abstractions, which have been used successfully in the classical setting. We show how to construct probability-aware Cartesian abstractions via a counterexample-guided abstraction refinement (CEGAR) loop akin to classical planning. This method is complete, meaning it guarantees convergence to the optimal expected cost if not terminated prematurely. Furthermore, we investigate the admissible combination of multiple such heuristics using saturated cost partitioning (SCP), marking its first application in the probabilistic setting. In our experiments, we show that probability-aware Cartesian abstractions yield much more informative heuristics than their determinization-based counterparts. Finally, we show that SCP yields probability-aware abstraction heuristics that are superior to the previous state of the art.

show abstract

Pattern Databases for Stochastic Shortest Path Problems

Cited by 3 publications

References 18 publications

Heuristic Search for Multi-Objective Probabilistic Planning

Heuristic Search for Multi-Objective Probabilistic Planning

A Theory of Merge-and-Shrink for Stochastic Shortest Path Problems

Cartesian Abstractions and Saturated Cost Partitioning in Probabilistic Planning

Contact Info

Product

Resources

About