Published March 8, 2021 | Version v3
Software Open

Proofs, Code and Data for Subset-Saturated Transition Cost Partitioning

  • 1. Linköping University
  • 2. University of Freiburg

Description

sstcp-code.zip contains an extended version of the Fast Downward planning system (http://fast-downward.org). Please see http://www.fast-downward.org for instructions on how to compile the planner. It contains additional code for optimal transition cost partitioning and subset-saturated transition cost partitioning.
The latest version is available at GitHub: https://github.com/drexlerd/fast-downward-sstcp

sstcp-single_order-experimental_data.zip contains experimental data for single orders of the heuristics.

sstcp-single_order-amask-experimental_data.zip contains experimental data for single orders of the heuristics and additional switching between operator and transition saturators depending on a simple threshold parameter that depends on the number of abstract transitions.

sstcp-multiple_orders-amask-experiment_data.zip contains experimental data for multiple orders of the heuristics and additional switching between operator and transition saturators depending on a simple threshold parameter that depends on the number of abstract transitions.

drexler-et-al-icaps2021-with-proof.pdf is the paper published at ICAPS2021 with proofs.
 

Notes

Other fundings: - Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by Knut and Alice Wallenberg Foundation - German Research Foundation (DFG) as part of the project EPSDAC (grant agreement no. MA 7790/1-1)

Files

drexler-et-al-icaps2021-with-proofs.pdf

Files (42.2 MB)

Name Size Download all
md5:24eaf01ebb28abaa37c4f4842855d3ae
280.0 kB Preview Download
md5:46a0c1b8de466febf1fb59c6e4d0baef
937.8 kB Preview Download
md5:91ab6970ed1abd26841ab7cab749ad2d
23.6 MB Preview Download
md5:8ca78e1b1caa6b858a7aa30e46cf02a4
7.2 MB Preview Download
md5:73d692e2f4ca9de5870af125ec5c3cdf
10.2 MB Preview Download

Additional details

Related works

Is new version of
Software: 10.5281/zenodo.4065414 (DOI)

Funding

TAILOR – Foundations of Trustworthy AI - Integrating Reasoning, Learning and Optimization 952215
European Commission
BDE – Beyond Distance Estimates: A New Theory of Heuristics for State-Space Search 817639
European Commission