Learning Portfolios of Automatically Tuned Planners

Seipp, Jendrik; Braun, Manuel; Garimort, Johannes; Helmert, Malte

doi:10.1609/icaps.v22i1.13538

Cited by 24 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fast Downward has many configurations; we use the same default configuration as in previous work (Fawcett et al 2011 Table 2: Overview of the planners in our experiments. Seipp et al 2012). We consider three versions of FF: the standard FF v2.3 (Hoffmann and Nebel 2001), FF-X which supports derived predicates (Thiébaux, Hoffmann, and Nebel 2005), and Metric-FF which supports action costs (Hoffmann 2003).…”

Section: Experiments Design and Methodologymentioning

confidence: 99%

Improved Features for Runtime Prediction of Domain-Independent Planners

Fawcett

Vallati

Hutter

et al. 2014

ICAPS

View full text Add to dashboard Cite

State-of-the-art planners often exhibit substantial runtime variation, making it useful to be able to efficiently predict how long a given planner will take to run on a given instance. In other areas of AI, such needs are met by building so-called empirical performance models (EPMs), statistical models derived from sets of problem instances and performance observations. Historically, such models have been less accurate for predicting the running times of planners. A key hurdle has been a relative weakness in instance features for characterizing instances: mappings from problem instances to real numbers that serve as the starting point for learning an EPM. We propose a new, extensive set of instance features for planning, and investigate its effectiveness across a range of model families. We built EPMs for various prominent planning systems on several thousand benchmark problems from the planning literature and from IPC benchmark sets, and conclude that our models predict runtime much more accurately than the previous state of the art. We also study the relative importance of these features.

show abstract

Section: Experiments Design and Methodologymentioning

confidence: 99%

Improved Features for Runtime Prediction of Domain-Independent Planners

Fawcett

Vallati

Hutter

et al. 2014

ICAPS

View full text Add to dashboard Cite

show abstract

“…Portfolios (Helmert, Röger, and Karpas 2011) statically select a set of planners to run in sequence, each with a set amount of timeout such that the sum of their timeout is equal to the total timeout of the portfolio planner. The partitioning of time resources for planners can be selected from training data or can simply be uniform (Seipp et al 2012). Similarly to portfolios, one may also just learn to choose a specific planner for a given domain (Fawcett et al 2011;Katz et al 2018;Ma et al 2020).…”

Section: Portfoliosmentioning

confidence: 99%

Novelty Heuristics, Multi-Queue Search, and Portfolios for Numeric Planning

Chen,

Thiébaux

2024

SOCS

View full text Add to dashboard Cite

Heuristic search is a powerful approach for solving planning problems and numeric planning is no exception. In this paper, we boost the performance of heuristic search for numeric planning with various powerful techniques orthogonal to improving heuristic informedness: numeric novelty heuristics, the Manhattan distance heuristic, and exploring the use of multi-queue search and portfolios for combining heuristics.

show abstract

“…We note that the design of Fast Downward-Autotune was explicitly inspired by an earlier version of our work described here, which was later presented at the Third Workshop on Planning and Learning (PAL 2011). More recently, a technique for statically configuring a portfolio of tuned planners was proposed (Seipp et al 2012). In this work, Seipp et al used ParamILS for tuning Fast Downward on several different well-known benchmark domains, combining the resulting configurations in a static domain-independent portfolio.…”

Section: Introductionmentioning

confidence: 99%

Automatic Generation of Efficient Domain-Optimized Planners from Generic Parametrized Planners

Vallati

Fawcett

Gerevini

et al. 2021

SOCS

View full text Add to dashboard Cite

When designing state-of-the-art, domain-independent planning systems, many decisions have to be made with respect to the domain analysis or compilation performed during preprocessing, the heuristic functions used during search, and other features of the search algorithm. These design decisions can have a large impact on the performance of the resulting planner. By providing many alternatives for these choices and exposing them as parameters, planning systems can in principle be configured to work well on different domains. However, planners are typically used in default configurations that have been chosen because of their good average performance over a set of benchmark domains, with limited experimentation over the potentially huge range of possible configurations. In this work, we propose a general framework for automatically configuring a parameterized planner, and show that substantial performance gains can be achieved. We apply the framework to the well-known LPG planner, which in the context of this work was expanded to 62 parameters and over 6.5 x 10^17 possible configurations. By using this highly parameterized planning system in combination with the state-of-the-art automatic algorithm configuration procedure ParamILS, excellent performance on a broad range of well-known benchmark domains was achieved, as also witnessed by the results of the learning track of the 7th International Planning Competition.

show abstract

Learning Portfolios of Automatically Tuned Planners

Cited by 24 publications

References 15 publications

Improved Features for Runtime Prediction of Domain-Independent Planners

Improved Features for Runtime Prediction of Domain-Independent Planners

Novelty Heuristics, Multi-Queue Search, and Portfolios for Numeric Planning

Automatic Generation of Efficient Domain-Optimized Planners from Generic Parametrized Planners

Contact Info

Product

Resources

About