2026-07-21T14:13:13Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:23884 2025-11-12T13:10:56Z ddc:004 open_access

RPID: Rust Programmable Interface for Domain-Independent Dynamic Programming Kuroiwa, Ryo Beck, J. Christopher Decision Diagrams & Dynamic Programming Modelling & Modelling Languages In domain-independent dynamic programming (DIDP), a problem is formulated as a dynamic programming (DP) model and then solved by a general-purpose solver. In the existing software for DIDP, a model is defined using expressions composed of a predefined set of operations. In this paper, we propose the Rust Programmable Interface for DIDP (RPID), new software for DIDP, where a model is defined by Rust functions. We discuss the design of RPID and compare it with existing DP-based frameworks, including decision diagram-based (DD-based) solvers. In our experiments, RPID is up to hundreds of times faster than the existing DIDP implementation with the same models. In addition, new DIDP models, enabled by the flexibility of RPID, outperform existing models in multiple problem classes. We also show that the relative performance of RPID and existing DD-based solvers depends on problem class with, so far, no clear dominant solver technology. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Ryo Kuroiwa and J. Christopher Beck 2025 Is Part Of LIPIcs, Volume 340, 31st International Conference on Principles and Practice of Constraint Programming (CP 2025) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.CP.2025.23 urn:nbn:de:0030-drops-238845 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.CP.2025.23 eng https://creativecommons.org/licenses/by/4.0/legalcode