Proof-carrying plans
Proof-carrying plans
It is becoming increasingly important to verify safety and security of AI applications. While declarative languages (of the kind found in automated planners and model checkers) are traditionally used for verifying AI systems, a big challenge is to design methods that generate verified executable programs. A good example of such a “verification to implementation” cycle is given by automated planning languages like PDDL, where plans are found via a model search in a declarative language, but then interpreted or compiled into executable code in an imperative language. In this paper, we show that this method can itself be verified. We present a formal framework and a prototype Agda implementation that represent PDDL plans as executable functions that inhabit types that are given by formulae describing planning problems. By exploiting the well-known Curry-Howard correspondence, type-checking then automatically ensures that the generated program corresponds precisely to the specification of the planning problem.
AI planning, Constructive logic, Curry-Howard correspondence, Dependent types, Verification
204-220
Schwaab, Christopher
82bdcdbf-4328-4db1-b6ca-3d5e31893022
Komendantskaya, Ekaterina
f12d9c23-5589-40b8-bcf9-a04fe9dedf61
Hill, Alasdair
e7a0d7d4-3106-4ab3-94bf-9fc8702bf42e
Farka, František
27fe3fde-d687-4882-a2a7-cae39126c589
Petrick, Ronald P.A.
031c87fc-f916-4e61-a77e-7002e5f53ad1
Wells, Joe
0242bfb6-faa4-49a9-b7e6-3acb219213c1
Hammond, Kevin
88dff6c4-1247-4fc0-bee5-b8aa994b8c30
1 January 2019
Schwaab, Christopher
82bdcdbf-4328-4db1-b6ca-3d5e31893022
Komendantskaya, Ekaterina
f12d9c23-5589-40b8-bcf9-a04fe9dedf61
Hill, Alasdair
e7a0d7d4-3106-4ab3-94bf-9fc8702bf42e
Farka, František
27fe3fde-d687-4882-a2a7-cae39126c589
Petrick, Ronald P.A.
031c87fc-f916-4e61-a77e-7002e5f53ad1
Wells, Joe
0242bfb6-faa4-49a9-b7e6-3acb219213c1
Hammond, Kevin
88dff6c4-1247-4fc0-bee5-b8aa994b8c30
Schwaab, Christopher, Komendantskaya, Ekaterina, Hill, Alasdair, Farka, František, Petrick, Ronald P.A., Wells, Joe and Hammond, Kevin
(2019)
Proof-carrying plans.
Alferes, José Júlio and Johansson, Moa
(eds.)
In Practical Aspects of Declarative Languages - 21st International Symposium, PADL 2019, Proceedings.
vol. 11372 LNCS,
Springer.
.
(doi:10.1007/978-3-030-05998-9_13).
Record type:
Conference or Workshop Item
(Paper)
Abstract
It is becoming increasingly important to verify safety and security of AI applications. While declarative languages (of the kind found in automated planners and model checkers) are traditionally used for verifying AI systems, a big challenge is to design methods that generate verified executable programs. A good example of such a “verification to implementation” cycle is given by automated planning languages like PDDL, where plans are found via a model search in a declarative language, but then interpreted or compiled into executable code in an imperative language. In this paper, we show that this method can itself be verified. We present a formal framework and a prototype Agda implementation that represent PDDL plans as executable functions that inhabit types that are given by formulae describing planning problems. By exploiting the well-known Curry-Howard correspondence, type-checking then automatically ensures that the generated program corresponds precisely to the specification of the planning problem.
This record has no associated files available for download.
More information
e-pub ahead of print date: 19 December 2018
Published date: 1 January 2019
Additional Information:
Publisher Copyright:
© Springer Nature Switzerland AG 2019.
Venue - Dates:
21st International Symposium on Practical Aspects of Declarative Languages, PADL 2019, , Lisbon, Portugal, 2019-01-14 - 2019-01-15
Keywords:
AI planning, Constructive logic, Curry-Howard correspondence, Dependent types, Verification
Identifiers
Local EPrints ID: 500428
URI: http://eprints.soton.ac.uk/id/eprint/500428
ISSN: 0302-9743
PURE UUID: 88a36c8c-c26a-4315-82a8-a6ff69e4e74d
Catalogue record
Date deposited: 29 Apr 2025 16:43
Last modified: 23 May 2025 02:08
Export record
Altmetrics
Contributors
Author:
Christopher Schwaab
Author:
Ekaterina Komendantskaya
Author:
Alasdair Hill
Author:
František Farka
Author:
Ronald P.A. Petrick
Author:
Joe Wells
Author:
Kevin Hammond
Editor:
José Júlio Alferes
Editor:
Moa Johansson
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics