Menu
▾
▴
potassco-users — Mailing list to ask general question about potassco tools
You can subscribe to this list here.
| 2010 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
(1) |
Oct
(8) |
Nov
|
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2011 |
Jan
|
Feb
|
Mar
(5) |
Apr
(4) |
May
(2) |
Jun
|
Jul
(7) |
Aug
(34) |
Sep
(11) |
Oct
(25) |
Nov
(9) |
Dec
(19) |
| 2012 |
Jan
(3) |
Feb
|
Mar
(7) |
Apr
(13) |
May
(13) |
Jun
(24) |
Jul
(16) |
Aug
(2) |
Sep
|
Oct
(14) |
Nov
(34) |
Dec
(9) |
| 2013 |
Jan
(38) |
Feb
(18) |
Mar
(17) |
Apr
(17) |
May
(36) |
Jun
(3) |
Jul
(4) |
Aug
(25) |
Sep
(25) |
Oct
(15) |
Nov
(29) |
Dec
(9) |
| 2014 |
Jan
(40) |
Feb
(84) |
Mar
(25) |
Apr
(47) |
May
(28) |
Jun
(21) |
Jul
(19) |
Aug
(27) |
Sep
(16) |
Oct
(68) |
Nov
(33) |
Dec
(15) |
| 2015 |
Jan
(15) |
Feb
(16) |
Mar
(2) |
Apr
(11) |
May
(16) |
Jun
(65) |
Jul
(26) |
Aug
(22) |
Sep
(18) |
Oct
(40) |
Nov
(14) |
Dec
(18) |
| 2016 |
Jan
(8) |
Feb
(38) |
Mar
(11) |
Apr
(47) |
May
(12) |
Jun
(5) |
Jul
(10) |
Aug
(3) |
Sep
(30) |
Oct
(32) |
Nov
(29) |
Dec
(10) |
| 2017 |
Jan
(5) |
Feb
(9) |
Mar
(38) |
Apr
(6) |
May
(2) |
Jun
(10) |
Jul
(26) |
Aug
(14) |
Sep
(5) |
Oct
(26) |
Nov
(15) |
Dec
(3) |
| 2018 |
Jan
(12) |
Feb
(4) |
Mar
(11) |
Apr
|
May
(11) |
Jun
(7) |
Jul
(7) |
Aug
(13) |
Sep
(10) |
Oct
(16) |
Nov
(21) |
Dec
(1) |
| 2019 |
Jan
(7) |
Feb
(12) |
Mar
(8) |
Apr
(17) |
May
(37) |
Jun
(19) |
Jul
(5) |
Aug
(10) |
Sep
(6) |
Oct
(6) |
Nov
(17) |
Dec
(12) |
| 2020 |
Jan
(10) |
Feb
(13) |
Mar
(25) |
Apr
(10) |
May
(13) |
Jun
(4) |
Jul
(11) |
Aug
(62) |
Sep
(2) |
Oct
(10) |
Nov
(9) |
Dec
(46) |
| 2021 |
Jan
(42) |
Feb
(28) |
Mar
(29) |
Apr
(28) |
May
(14) |
Jun
(14) |
Jul
(3) |
Aug
(29) |
Sep
(24) |
Oct
(36) |
Nov
(57) |
Dec
(18) |
| 2022 |
Jan
(5) |
Feb
(59) |
Mar
(13) |
Apr
(22) |
May
(60) |
Jun
(21) |
Jul
(35) |
Aug
(4) |
Sep
(5) |
Oct
(9) |
Nov
(13) |
Dec
(18) |
| 2023 |
Jan
(47) |
Feb
(26) |
Mar
(4) |
Apr
(29) |
May
(19) |
Jun
(9) |
Jul
(35) |
Aug
(13) |
Sep
(27) |
Oct
(10) |
Nov
(17) |
Dec
(7) |
| 2024 |
Jan
(14) |
Feb
(11) |
Mar
(9) |
Apr
(5) |
May
(6) |
Jun
(21) |
Jul
|
Aug
(3) |
Sep
(8) |
Oct
|
Nov
|
Dec
|
|
From: Grassauer, L. <luk...@st...> - 2024-09-10 12:36:40
|
Hey, I would also like to highlight tree-sitter https://tree-sitter.github.io/tree-sitter/ and https://github.com/potassco/tree-sitter-clingo which enables support for editors like Helix https://helix-editor.com/ All that is missing is a highlights.scm. But that should not be hard to add. Best, Lukas Grassauer ________________________________ From: Roland Kaminski <kam...@cs...> Sent: Tuesday, September 10, 2024 1:14:16 PM To: pot...@li... <pot...@li...> Subject: Re: [Potassco-users] ASP/Clingo syntax highlighting on github repos Hi, it would probably require some effort to get proper syntax highlighting support on github. Having support in editors is in my opinion more important. I list below two projects I have been working on. The first is for highlighting in vim/neovim and the second for highlighting with pygments (which can for example be used with the latex minted package). - https://github.com/rkaminsk/vim-syntax-clingo - https://github.com/rkaminsk/pygments_clingo There are also more such projects for other editors: - https://github.com/llaisdy/clingo-mode - https://github.com/ArnaudBelcour/asp-syntax-highlight -R On Sunday 8 September 2024 18:49:02 CEST Bertram Ludaescher wrote: > Hi everyone, > > I was looking into ways to get syntax highlighting for ASP/clingo programs > on github, so that when one browses an ASP program on a github repo it > looks nicer / is more readable. > > One workaround is to pretend that ASP files (extension = ".lp" ) follow > Prolog syntax and create a ".gitattributes" file with the following > content: > *.lp linguist-language=Prolog > > By adding this file to a repo, all ".lp" files will get Prolog syntax > highlighting. > To test this, I cloned the clingo user guide and added > https://github.com/ludaesch/clingo_guide/blob/master/.gitattributes > > Then all example files get some highlighting. Such a quick hack ;-) > For example, this tiny example looks reasonable: > https://github.com/ludaesch/clingo_guide/blob/master/examples/fly.lp > > Even rules with clingo-specific syntax come out OK: > https://github.com/ludaesch/clingo_guide/blob/master/examples/aggr.lp > > But of course many clingo-specific constructs aren't recognized and thus > not highlighted correctly. Has anyone looked into doing this? > > Some info on how this might be done is here: > > https://github.com/github-linguist/linguist/blob/master/docs/overrides.md > > Thanks, cheers, > > Bertram > > *Bertram Ludäscher* > Professor, School of Information Sciences (iSchool@Illinois > <http://ischool.illinois.edu/>) > Director, Center for Informatics Research in Science & Scholarship (CIRSS) > Faculty Affiliate, NCSA <http://www.ncsa.illinois.edu/> & Computer Science > <https://siebelschool.illinois.edu/> > *University of Illinois at Urbana-Champaign* > lud...@il... |
|
From: Roland K. <kam...@cs...> - 2024-09-10 11:09:28
|
Hi, it would probably require some effort to get proper syntax highlighting support on github. Having support in editors is in my opinion more important. I list below two projects I have been working on. The first is for highlighting in vim/neovim and the second for highlighting with pygments (which can for example be used with the latex minted package). - https://github.com/rkaminsk/vim-syntax-clingo - https://github.com/rkaminsk/pygments_clingo There are also more such projects for other editors: - https://github.com/llaisdy/clingo-mode - https://github.com/ArnaudBelcour/asp-syntax-highlight -R On Sunday 8 September 2024 18:49:02 CEST Bertram Ludaescher wrote: > Hi everyone, > > I was looking into ways to get syntax highlighting for ASP/clingo programs > on github, so that when one browses an ASP program on a github repo it > looks nicer / is more readable. > > One workaround is to pretend that ASP files (extension = ".lp" ) follow > Prolog syntax and create a ".gitattributes" file with the following > content: > *.lp linguist-language=Prolog > > By adding this file to a repo, all ".lp" files will get Prolog syntax > highlighting. > To test this, I cloned the clingo user guide and added > https://github.com/ludaesch/clingo_guide/blob/master/.gitattributes > > Then all example files get some highlighting. Such a quick hack ;-) > For example, this tiny example looks reasonable: > https://github.com/ludaesch/clingo_guide/blob/master/examples/fly.lp > > Even rules with clingo-specific syntax come out OK: > https://github.com/ludaesch/clingo_guide/blob/master/examples/aggr.lp > > But of course many clingo-specific constructs aren't recognized and thus > not highlighted correctly. Has anyone looked into doing this? > > Some info on how this might be done is here: > > https://github.com/github-linguist/linguist/blob/master/docs/overrides.md > > Thanks, cheers, > > Bertram > > *Bertram Ludäscher* > Professor, School of Information Sciences (iSchool@Illinois > <http://ischool.illinois.edu/>) > Director, Center for Informatics Research in Science & Scholarship (CIRSS) > Faculty Affiliate, NCSA <http://www.ncsa.illinois.edu/> & Computer Science > <https://siebelschool.illinois.edu/> > *University of Illinois at Urbana-Champaign* > lud...@il... |
|
From: Bertram L. <lud...@il...> - 2024-09-08 17:54:17
|
Hi everyone, I was looking into ways to get syntax highlighting for ASP/clingo programs on github, so that when one browses an ASP program on a github repo it looks nicer / is more readable. One workaround is to pretend that ASP files (extension = ".lp" ) follow Prolog syntax and create a ".gitattributes" file with the following content: *.lp linguist-language=Prolog By adding this file to a repo, all ".lp" files will get Prolog syntax highlighting. To test this, I cloned the clingo user guide and added https://github.com/ludaesch/clingo_guide/blob/master/.gitattributes Then all example files get some highlighting. Such a quick hack ;-) For example, this tiny example looks reasonable: https://github.com/ludaesch/clingo_guide/blob/master/examples/fly.lp Even rules with clingo-specific syntax come out OK: https://github.com/ludaesch/clingo_guide/blob/master/examples/aggr.lp But of course many clingo-specific constructs aren't recognized and thus not highlighted correctly. Has anyone looked into doing this? Some info on how this might be done is here: https://github.com/github-linguist/linguist/blob/master/docs/overrides.md Thanks, cheers, Bertram *Bertram Ludäscher* Professor, School of Information Sciences (iSchool@Illinois <http://ischool.illinois.edu/>) Director, Center for Informatics Research in Science & Scholarship (CIRSS) Faculty Affiliate, NCSA <http://www.ncsa.illinois.edu/> & Computer Science <https://siebelschool.illinois.edu/> *University of Illinois at Urbana-Champaign* lud...@il... |
|
From: Susana H. M. L. <hah...@un...> - 2024-09-02 18:08:20
|
Hi everyone, We are excited to announce a new version of Clinguin, our system for creating User Interfaces using only ASP. Here are the important links: Improved Documentation <https://clinguin.readthedocs.io/en/latest/> Source Code on GitHub <https://github.com/potassco/clinguin> For those who have been using Clinguin version 1, please check out the release notes for version 2.0.0 <https://github.com/potassco/clinguin/releases/tag/v2.0.0>. Please note that there are some backward-incompatible changes, so you might need to make minor adjustments to your existing UI encodings. We welcome any feedback, as it will help us improve further. Please share your thoughts and suggestions through GitHub issues. Best regards, The Potassco Team PS: There will be a presentation on Clinguin in the weekly seminar slot on September 11th at 3 PM ECT. |
|
From: Lukas L. <luk...@de...> - 2024-09-02 08:10:09
|
Hello everyone, I wanted to naively ask if there is a way to estimate the number of answer sets of a clingo program. For a project, I want to change a relatively large program, but keep the number of answer sets similar in size. I found this repository https://github.com/meelgroup/ApproxASP2 and the corresponding paper, but the repository seems to be no longer maintained and I can't get the code to run. Do you know of other ways for model counting in clingo programs? Best regards Lukas |
|
From: Tarzariol, A. <Ali...@aa...> - 2024-08-20 10:42:02
|
===================================================================== CALL FOR PAPERS TAASP 2024 Workshop on Trends and Applications of Answer Set Programming November 25-26, 2024 https://taasp.at/2024/ ===================================================================== AIMS AND SCOPE Answer set programming (ASP) is a widely used declarative programming paradigm based on the stable-model semantics. Over the years, the formalism has been extended with new language elements, advanced evaluation techniques have been developed, and efficient and expressive systems have been realized. These advancements turned ASP into a rich modeling language that was successfully applied to various tasks in artificial intelligence and beyond, including combinatorial problems and knowledge-intense tasks. Driven by the needs of applications, ongoing research in the field comprises, for instance, the integration with other paradigms (such as constraint programming), distributed computation, temporal and stream reasoning. These developments aim at the ultimate goal of equipping ASP with the features needed for realizing modern information systems. This workshop focuses on recent trends in the area of ASP, including novel techniques for the evaluation of ASP programs, extensions of the formalism, integration with other formalisms, advancements of systems, and benchmark suites. Furthermore, discussions of applications of ASP and its extensions in AI and industry are welcome. SUBMISSIONS We welcome two categories of submissions: * full papers presenting original research (at most 15 pages plus references and optional appendix), and * extended abstracts of already published or preliminary research (at most 2 pages plus references and optional appendix). All submissions should be in the Springer LNCS format: https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines Paper submission will be handled electronically by means of the Easychair system. All submissions will be peer-reviewed. The submission page is available here: https://easychair.org/conferences/?conf=taasp2024 IMPORTANT DATES Paper submission deadline: October 4, 2024 Author notification: November 4, 2024 Camera-ready articles due: November 18, 2024 Workshop: November 25-26, 2024 LOCATION The workshop will be held at Alpen-Adria-Universitaet Klagenfurt, Austria. PROCEEDINGS TAASP is a non-archival venue and there will be no published proceedings. However, informal proceedings will be provided and the papers will be posted informally on the workshop website. Therefore, it will be possible to submit to other conferences and journals both in parallel and subsequent to TAASP 2024. ORGANIZATION COMMITTEE Alice Tarzariol, Alpen-Adria-Universitaet Klagenfurt, Austria Klaus Strauch, Universitaet Potsdam, Germany Tobias Geibinger, Technische Universitaet Wien, Austria CONTACT Workshop page: https://taasp.at/2024/ Submission page: https://easychair.org/conferences/?conf=taasp2024 Email address: taasp2024(at)easychair.org |
|
From: Roland K. <kam...@cs...> - 2024-08-05 09:28:44
|
Hi, I do not know the memory requirements of your application. Some things to
note:
- wasm is limited to 4GB as it uses 32bit addresses.
- if you are running in a browser, stricter memory requirements might apply
Native versions of clingo can access much more memory. -R
On Monday 5 August 2024 10:52:08 CEST Patrick Galea wrote:
> Hello!
>
> I am trying to convert Chess into an ASP program. I have previously run
> into issues regarding grounding where a 'heap allocation' error has
> occurred. However, I am now running into an issue where the output is
> simply:
>
> { Result: 'ERROR', Error: 'Aborted()' }
>
> For reference, I am using the clingo-wasm package
> https://github.com/domoritz/clingo-wasm and am curious if anyone knows
> anything about this error? Alternatively, is there any way to increase
> memory of clingo?
>
> I am happy to provide more information if anyone needs!
>
> Thanks
|
|
From: Patrick G. <pat...@gm...> - 2024-08-05 08:52:26
|
Hello!
I am trying to convert Chess into an ASP program. I have previously run
into issues regarding grounding where a 'heap allocation' error has
occurred. However, I am now running into an issue where the output is
simply:
{ Result: 'ERROR', Error: 'Aborted()' }
For reference, I am using the clingo-wasm package
https://github.com/domoritz/clingo-wasm and am curious if anyone knows
anything about this error? Alternatively, is there any way to increase
memory of clingo?
I am happy to provide more information if anyone needs!
Thanks
|
|
From: Gerhard F. <ger...@gm...> - 2024-06-24 13:54:55
|
Project researcher predoctoral (all genders welcome) Department of Artificial Intelligence and Cybersecurity University of Klagenfurt, Austria Scientific Staff | Full time Application deadline: 01 August 2024 Announcement The University of Klagenfurt, with approximately 1,500 employees and over 12,000 students, is located in the Alps-Adriatic region and consistently achieves excellent placements in rankings. The motto “per aspera ad astra” underscores our firm commitment to the pursuit of excellence in all activities in research, teaching, and university management. The principles of equality, diversity, health, sustainability, and compatibility of work and family life serve as the foundation for our work at the university. The University of Klagenfurt is pleased to announce the following open position at the Department of Artificial Intelligence and Cybersecurity, Intelligent Systems and Business Informatics, at the Faculty of Technical Sciences with an expected starting date of September 1, 2024: Project researcher predoctoral (all genders welcome) Your personal sphere of influence As a project researcher, you will join the research team around Prof. Gerhard Friedrich and Prof. Konstantin Schekotihin working on the project “SAving Energy by Learning and ImproviNG logic-based optimization models (SAELING)” funded by the Austrian Research Promotion Agency in the focus area “Artificial Intelligence for Green.” You will work in a multilingual research team with Siemens, Voestalpine, and KU Leuven, Belgium (Prof. Tias Guns). The SAELING research project will use machine learning methods to generate prediction models that will be used for scheduling in multi-objective optimization problems to find energy- and resource-saving strategies for the industrial use case of metal processing. This project combines sub-symbolic and symbolic methods of artificial intelligence and, therefore, is closely connected to the Austrian cluster of excellence “Bilateral Artificial Intelligence,” in which the Department of Artificial Intelligence and Cybersecurity participates. Clusters of excellence are the Austrian flagships of basic research. If you are open to new challenges and value teamwork in a successful research environment, you are welcome to join our team! Level of employment: 100 % (40 hours/week) Minimum salary: € 50,103.20 per annum (gross); classification according to collective agreement: B1 Limited to: 3 years Application deadline: by August 1, 2024 Area of responsibility Tasks and responsibilities: Autonomous scientific work, including the publication of research articles in the field of artificial intelligence, with a specific emphasis on symbolic AI, in particular logic, logic programming, and multi-objective optimization and its integration with machine learning based on artificial neural networks. Conduct theoretical and practical (empirical) research Aim to submit a dissertation and acquire a doctoral degree Collaboration with industry partners Specification of application scenarios Implementation and evaluation of proof-of-concept prototypes Contribution to organizational and administrative tasks Participation in public relations activities The successful candidate is expected to pursue a PhD within the thematic doctoral program Informatics at the University of Klagenfurt. This position serves the purposes of the vocational and scientific education of graduates of Master’s or Diploma degree programs with the goal of completing a doctoral degree in Technical Sciences. Applications by persons who have already completed a subject-specific doctoral degree can, therefore, not be considered. In research and teaching, the research group Intelligent Systems and Business Informatics works on knowledge representation and reasoning in intelligent systems with an emphasis on application in production. The research group focuses on logic-based/constrained-based problem-solvers, their integration with machine learning, and their applicability to large practical problem instances. The spectrum of the research activities ranges from developing new problem-solving methods in declarative languages and their combination with machine learning to the implementation of systems supporting such methods. Requirements Prerequisites for the appointment: Master’s degree at a domestic or foreign higher education institution in computer science or a related field. This requirement must be fulfilled two weeks before the starting date at the latest; hence, the last possible deadline for meeting this requirement is August 18, 2024 Background in one or more of the following fields: symbolic artificial intelligence, such as logic programming, constraint programming, deep learning, or reinforcement learning Solid communication and dissemination skills Fluency in English (both written and spoken) Desired skills Ability to work in a diverse team Programming skills, preferably Python Experience with answer set or constraint programming tools, e.g., clingo or OR-Tools Knowledge of deep learning frameworks, like PyTorch or TensorFlow German language skills are not required, but successful applicants will be expected to acquire them during their appointment. Additional information Our offer The employment contract is concluded for the position as project researcher (predoctoral) and stipulates a starting salary of € 3,578.80 gross per month (14 times a year). The University of Klagenfurt also offers: Personal and professional advanced training courses, management and career coaching, including bespoke training for women in science Numerous attractive additional benefits, see also https://jobs.aau.at/en/the-university-as-employer/ Diversity- and family-friendly university culture The opportunity to live and work in the attractive Alps-Adriatic region with a wide range of leisure activities in the spheres of culture, nature and sports The application: If you are interested in this position, please apply in English providing the following documents: Letter of application explaining the motivation and including a statement of interest in research Curriculum vitae (please do not include a photo) Copies of degree certificates (Bachelor and Master) Copies of official transcripts (Bachelor and Master) containing the list of all courses and grades Master’s thesis. If the thesis is not available, the candidate should provide an explanation. Three references (contact details of persons who the university may contact by email or telephone for information purposes) To apply, please send your application to Petra Wiesner (pet...@aa... <mailto:pet...@aa...>). If an applicant has not received the master’s degree by the application deadline, the applicant should provide a declaration, written either by a supervisor or by the candidate themselves, on the feasibility of finishing the master’s degree before August 18, 2024. Candidates must provide proof that they meet the required qualifications by August 18, 2024, at the latest. For further information on this specific vacancy, please contact Prof. Dr. Gerhard Friedrich (ger...@aa... <mailto:ger...@aa...>). General information about the university as an employer can be found at https://jobs.aau.at/en/the-university-as-employer/. The University of Klagenfurt aims to increase the proportion of women and therefore specifically invites qualified women to apply for the position. Where the qualification is equivalent, women will be given preferential consideration. People with disabilities or chronic diseases, who fulfill the requirements, are particularly encouraged to apply. Travel and accommodation costs incurred during the application process will not be refunded. Translations into other languages shall serve informational purposes only. This job offer can also be found at https://jobboerse.aau.at/job/project-researcher-predoctoral-all-genders-welcome/. |
|
From: Andrew C. <and...@cs...> - 2024-06-21 13:37:07
|
Thanks! Kind regards, Andrew > On 21 Jun 2024, at 14:30, Benjamin Kaufmann <kau...@cs...> wrote: > > Hi Andrew, > >> Here is a simpel failing example: >> […] > > Thanks. This is indeed an issue in clasp's SAT-preprocessor, which eagerly eliminates the size/1 atoms from the program and hence the domain heuristic can no longer decide on them. > > On the solver side, one trivial fix would be to always exclude atoms for which there is a heuristic modification from variable elimination. > > I'll address this issue in the next release. For now, I would suggest not using sat-preprocessing together with the domain heuristic. > > Best regards, > Ben |
|
From: Benjamin K. <kau...@cs...> - 2024-06-21 13:30:29
|
Hi Andrew, > Here is a simpel failing example: > […] Thanks. This is indeed an issue in clasp's SAT-preprocessor, which eagerly eliminates the size/1 atoms from the program and hence the domain heuristic can no longer decide on them. On the solver side, one trivial fix would be to always exclude atoms for which there is a heuristic modification from variable elimination. I'll address this issue in the next release. For now, I would suggest not using sat-preprocessing together with the domain heuristic. Best regards, Ben |
|
From: Andrew C. <and...@cs...> - 2024-06-21 11:35:44
|
Hi Ben,
Here is a simpel failing example:
```
{p(1..10)}.
size(K):-
#count{A : p(A)} == K.
#heuristic size(N). [1000-N,true]
```
Working as I expected:
```
clingo bug.pl --heuristic=Domain -n10
clingo version 5.7.1
Reading from bug.pl
Solving...
Answer: 1
size(0)
Answer: 2
p(3) size(1)
Answer: 3
p(5) size(1)
Answer: 4
p(6) size(1)
Answer: 5
p(2) size(1)
Answer: 6
p(10) size(1)
Answer: 7
p(9) size(1)
Answer: 8
p(8) size(1)
Answer: 9
p(7) size(1)
Answer: 10
p(1) size(1)
SATISFIABLE
```
Not working as I expected:
```
clingo bug.pl --heuristic=Domain -n10 --sat-pre=1
clingo version 5.7.1
Reading from bug.pl
Solving...
Answer: 1
size(0)
Answer: 2
p(7) size(1)
Answer: 3
p(6) size(1)
Answer: 4
p(6) p(7) size(2)
Answer: 5
p(3) size(1)
Answer: 6
p(3) p(6) size(2)
Answer: 7
p(3) p(7) size(2)
Answer: 8
p(3) p(6) p(7) size(3)
Answer: 9
p(10) size(1)
Answer: 10
p(7) p(10) size(2)
SATISFIABLE
```
Kind regards,
Andrew
> On 21 Jun 2024, at 12:10, Benjamin Kaufmann <kau...@cs...> wrote:
>
> Hi Andrew,
>
> I would expect this to work or at least produce a warning. But maybe there is an issue with preprocessing when heuristic modifications are subject to conditions.
>
> Could you provide a (preferably minimal) example program?
>
> Best regards,
> Ben
>
>> Am 21.06.2024 um 12:53 schrieb Andrew Cropper via Potassco-users <pot...@li...>:
>>
>> Hi,
>>
>> We use Clingo and its domain heuristics like so `clingo problem.pl --heuristic=Domain`. Everything works as expected and Clingo returns models in the order specified by the domain heuristic.
>>
>> I tried enabling the "Run SatELite-like preprocessing” command like so `clingo problem.pl --heuristic=Domain —sat-prep=1`. However, now Clingo ignores the domain heuristic and does not return models in the expected order.
>>
>> Is this issue a bug, or am I misunderstanding the impact of the preprocessing step?
>>
>> Kind regards,
>>
>> Andrew
>>
>>
>>
>> _______________________________________________
>> Potassco-users mailing list
>> Pot...@li...
>> https://lists.sourceforge.net/lists/listinfo/potassco-users
>
|
|
From: Benjamin K. <kau...@cs...> - 2024-06-21 11:11:19
|
Hi Andrew, I would expect this to work or at least produce a warning. But maybe there is an issue with preprocessing when heuristic modifications are subject to conditions. Could you provide a (preferably minimal) example program? Best regards, Ben > Am 21.06.2024 um 12:53 schrieb Andrew Cropper via Potassco-users <pot...@li...>: > > Hi, > > We use Clingo and its domain heuristics like so `clingo problem.pl --heuristic=Domain`. Everything works as expected and Clingo returns models in the order specified by the domain heuristic. > > I tried enabling the "Run SatELite-like preprocessing” command like so `clingo problem.pl --heuristic=Domain —sat-prep=1`. However, now Clingo ignores the domain heuristic and does not return models in the expected order. > > Is this issue a bug, or am I misunderstanding the impact of the preprocessing step? > > Kind regards, > > Andrew > > > > _______________________________________________ > Potassco-users mailing list > Pot...@li... > https://lists.sourceforge.net/lists/listinfo/potassco-users |
|
From: Andrew C. <and...@cs...> - 2024-06-20 13:41:31
|
Hi, We use Clingo and its domain heuristics like so `clingo problem.pl --heuristic=Domain`. Everything works as expected and Clingo returns models in the order specified by the domain heuristic. I tried enabling the "Run SatELite-like preprocessing” command like so `clingo problem.pl --heuristic=Domain —sat-prep=1`. However, now Clingo ignores the domain heuristic and does not return models in the expected order. Is this issue a bug, or am I misunderstanding the impact of the preprocessing step? Kind regards, Andrew |
|
From: Zhiliang X. <Xi...@ca...> - 2024-06-14 16:08:11
|
I see. Thanks a lot Roland! -Zhiliang ________________________________ From: Roland Kaminski <kam...@cs...> Sent: 14 June 2024 16:32 To: pot...@li... <pot...@li...> Subject: Re: [Potassco-users] Computing cautious answer set of multiple superset optimal models Hi, solving disjunctive logic programs raises complexity to the second level of the polynomial hierarchy. As I tried to hint at, this approach will most likely not scale to large instances. -R On Friday 14 June 2024 16:34:56 CEST Zhiliang Xiang wrote: > Thank you, Roland. > > Could you please elaborate on what you mean by "However, this relies on a > reduction to disjunctive logic program. As such it will come with a hefty > performance penalty. "? > > This seems important to understand, as we tend to encounter scalability > issues when the dataset gets larger, even when using Asprin. > > Best regards, > > Zhiliang > > > ________________________________ > From: Roland Kaminski <kam...@cs...> > Sent: 13 June 2024 19:25 > To: pot...@li... > <pot...@li...> Subject: Re: [Potassco-users] > Computing cautious answer set of multiple superset optimal models > > Hi, approaching this via meta programming would work. The metasp project can > easily be adjusted to support a subset maximization certeria. It would even > work out of the box with the cautious consequences option. However, this > relies on a reduction to disjunctive logic program. As such it will come > with a hefty performance penalty. I am not sure it will scale to your use > case. -R > On Thursday 13 June 2024 20:02:35 CEST Zhiliang Xiang wrote: > > Hi Roland, > > If I understood correctly, you are suggesting adopting an enumeration > > scheme similar to how Clingo enumerates cautious consequences. If that's > > the case, I suppose it would likely involve modifications on Asprin's > > source code on enumeration, I'm unsure where to begin and how to do this > > non-intrusively. Do you have any suggestions in this regard? In fact, one > > of my colleagues recommended trying meta-programming [1] for > > optimisation, then combining it with cautious reasoning mode. What do you > > think? Would this be a plausible approach? Thank you very much, > > Zhiliang > > > > [1] > > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.cs.u<https://www.cs.u/> > > ni-potsdam.de%2Fwv%2Fpublications%2FTEMP_journals%2Ftplp%2FGebserKS11&data > > =05%7C02%7Cxiangz6%40cardiff.ac.uk%7C3b135f868fb449af6e1708dc8bd6f273%7Cbd > > b74b3095684856bdbf06759778fcbc%7C1%7C0%7C638539002515068396%7CUnknown%7CTW > > FpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0% > > 3D%7C0%7C%7C%7C&sdata=JD0hdb13V8iFPuNHL%2Ftnn1rKfWToAAUp6Zo1MSw4CrU%3D&res > > erved=0<https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.cs.uni-potsdam.de%2Fwv%2Fpublications%2FTEMP_journals%2Ftplp%2FG&data=05%7C02%7Cxiangz6%40cardiff.ac.uk%7C6bca233b567d4d4c5f2008dc8c87f075%7Cbdb74b3095684856bdbf06759778fcbc%7C1%7C0%7C638539762689426975%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=IFgYvXJE8uln5JLmILfIwNgHKvNIU2w2L06xFyaC7qU%3D&reserved=0 > > ebserKS11> x.pdf ________________________________ > > From: Roland Kaminski <kam...@cs...> > > Sent: 13 June 2024 17:04 > > To: pot...@li... > > <pot...@li...>; Javier Romero Davila > > <jav...@un...> Subject: Re: [Potassco-users] > > [Extern] Computing cautious answer set of multiple superset optimal models > > > > Maybe you can refine the enumeration scheme? If I am not mistaken, you > > can still gather a set of atoms that is not part of the cautious > > consequences. This set gets larger with each model found. We can add > > constraints pruning all models that are subsets of this set. These > > constraints should not interfere with the subset maximization. Maybe > > this would help... -R > > > > On Thu, 2024-06-13 at 15:22 +0000, Zhiliang Xiang wrote: > > > Hi Javier, > > > Thank you for your reply. > > > I forgot to mention that I did try to enumerate all the optimal > > > models, but unfortunately, the process did not terminate even after > > > days, hitting timeout or memory limits. > > > Given this, do you have any other suggestions or know of any other > > > techniques to approach this problem? > > > Best regards, > > > Zhiliang > > > From: Javier Romero Davila <jav...@un...> > > > Sent: 13 June 2024 05:39 > > > To: pot...@li... < > > > pot...@li...>; Zhiliang Xiang < > > > Xi...@ca...> > > > Subject: Re: [Extern][Potassco-users] Computing cautious answer set > > > of multiple superset optimal models > > > > > > External email to Cardiff University - Take care when > > > replying/opening attachments or links. > > > Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth > > > ateb/agor atodiadau neu ddolenni. > > > > > > > > > > > > Hi Zhliang, > > > > > > Asprin does not have an option to compute cautious consequences wrt > > > optimal > > > models. > > > You would have to use asprin to compute all optimal models, > > > and then compute the intersection yourself. > > > > > > Best wishes, > > > Javier > > > > > > On Wed, 12 Jun 2024 17:11:20 +0000 > > > > > > Zhiliang Xiang <Xi...@ca...> wrote: > > > > Dear Community, > > > > I am currently working on a project that involves computing the > > > > > > > >cautious answer set of multiple superset optimal models with respect > > > >to certain objective atoms. > > > > > > > > I attempted to enable cautious reasoning mode by passing the > > > > > > solving > > > > > > >option --enum-mode=cautious to the Clingo Control object used by > > > >Asprin, where a superset preference is specified. However, this did > > > >not produce the desired effect. > > > > > > > > Does anyone know how to enable cautious reasoning mode when using > > > > > > > >Asprin? If this is not possible, are there any alternative > > > > > > approaches > > > > > > >to achieve this? > > > > > > > > Thank you, > > > > Zhiliang > > > > > > _______________________________________________ > > > Potassco-users mailing list > > > Pot...@li... > > > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists<https://lists/>. > > > so<https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.so%2F&data=05%7C02%7Cxiangz6%40cardiff.ac.uk%7C6bca233b567d4d4c5f2008dc8c87f075%7Cbdb74b3095684856bdbf06759778fcbc%7C1%7C0%7C638539762689441877%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=laNjWpjEo5CXBA62a5xL5ZZ%2FzgF43oAmwiv9Q6XWIr0%3D&reserved=0<https://lists.so/>> > > > urceforge.net%2Flists%2Flistinfo%2Fpotassco-users&data=05%7C02%7Cxiangz > > > 6%4 > > > 0cardiff.ac.uk%7Cdb0bfcd8ae3b4e0c2f6308dc8bc2ca4f%7Cbdb74b3095684856bdb > > > f06 > > > 759778fcbc%7C1%7C0%7C638538915931326095%7CUnknown%7CTWFpbGZsb3d8eyJWIjo > > > iMC > > > 4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&s > > > dat > > > a=zj2R2qpv%2FzN4%2FBaTCdLL6hFAAyn4O9t1%2BaDqj29HDj0%3D&reserved=0<https > > > :// lists.sourceforge.net/lists/listinfo/potassco-users> |
|
From: Roland K. <kam...@cs...> - 2024-06-14 15:35:55
|
Hi, solving disjunctive logic programs raises complexity to the second level of the polynomial hierarchy. As I tried to hint at, this approach will most likely not scale to large instances. -R On Friday 14 June 2024 16:34:56 CEST Zhiliang Xiang wrote: > Thank you, Roland. > > Could you please elaborate on what you mean by "However, this relies on a > reduction to disjunctive logic program. As such it will come with a hefty > performance penalty. "? > > This seems important to understand, as we tend to encounter scalability > issues when the dataset gets larger, even when using Asprin. > > Best regards, > > Zhiliang > > > ________________________________ > From: Roland Kaminski <kam...@cs...> > Sent: 13 June 2024 19:25 > To: pot...@li... > <pot...@li...> Subject: Re: [Potassco-users] > Computing cautious answer set of multiple superset optimal models > > Hi, approaching this via meta programming would work. The metasp project can > easily be adjusted to support a subset maximization certeria. It would even > work out of the box with the cautious consequences option. However, this > relies on a reduction to disjunctive logic program. As such it will come > with a hefty performance penalty. I am not sure it will scale to your use > case. -R > On Thursday 13 June 2024 20:02:35 CEST Zhiliang Xiang wrote: > > Hi Roland, > > If I understood correctly, you are suggesting adopting an enumeration > > scheme similar to how Clingo enumerates cautious consequences. If that's > > the case, I suppose it would likely involve modifications on Asprin's > > source code on enumeration, I'm unsure where to begin and how to do this > > non-intrusively. Do you have any suggestions in this regard? In fact, one > > of my colleagues recommended trying meta-programming [1] for > > optimisation, then combining it with cautious reasoning mode. What do you > > think? Would this be a plausible approach? Thank you very much, > > Zhiliang > > > > [1] > > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.cs.u > > ni-potsdam.de%2Fwv%2Fpublications%2FTEMP_journals%2Ftplp%2FGebserKS11&data > > =05%7C02%7Cxiangz6%40cardiff.ac.uk%7C3b135f868fb449af6e1708dc8bd6f273%7Cbd > > b74b3095684856bdbf06759778fcbc%7C1%7C0%7C638539002515068396%7CUnknown%7CTW > > FpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0% > > 3D%7C0%7C%7C%7C&sdata=JD0hdb13V8iFPuNHL%2Ftnn1rKfWToAAUp6Zo1MSw4CrU%3D&res > > erved=0<https://www.cs.uni-potsdam.de/wv/publications/TEMP_journals/tplp/G > > ebserKS11> x.pdf ________________________________ > > From: Roland Kaminski <kam...@cs...> > > Sent: 13 June 2024 17:04 > > To: pot...@li... > > <pot...@li...>; Javier Romero Davila > > <jav...@un...> Subject: Re: [Potassco-users] > > [Extern] Computing cautious answer set of multiple superset optimal models > > > > Maybe you can refine the enumeration scheme? If I am not mistaken, you > > can still gather a set of atoms that is not part of the cautious > > consequences. This set gets larger with each model found. We can add > > constraints pruning all models that are subsets of this set. These > > constraints should not interfere with the subset maximization. Maybe > > this would help... -R > > > > On Thu, 2024-06-13 at 15:22 +0000, Zhiliang Xiang wrote: > > > Hi Javier, > > > Thank you for your reply. > > > I forgot to mention that I did try to enumerate all the optimal > > > models, but unfortunately, the process did not terminate even after > > > days, hitting timeout or memory limits. > > > Given this, do you have any other suggestions or know of any other > > > techniques to approach this problem? > > > Best regards, > > > Zhiliang > > > From: Javier Romero Davila <jav...@un...> > > > Sent: 13 June 2024 05:39 > > > To: pot...@li... < > > > pot...@li...>; Zhiliang Xiang < > > > Xi...@ca...> > > > Subject: Re: [Extern][Potassco-users] Computing cautious answer set > > > of multiple superset optimal models > > > > > > External email to Cardiff University - Take care when > > > replying/opening attachments or links. > > > Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth > > > ateb/agor atodiadau neu ddolenni. > > > > > > > > > > > > Hi Zhliang, > > > > > > Asprin does not have an option to compute cautious consequences wrt > > > optimal > > > models. > > > You would have to use asprin to compute all optimal models, > > > and then compute the intersection yourself. > > > > > > Best wishes, > > > Javier > > > > > > On Wed, 12 Jun 2024 17:11:20 +0000 > > > > > > Zhiliang Xiang <Xi...@ca...> wrote: > > > > Dear Community, > > > > I am currently working on a project that involves computing the > > > > > > > >cautious answer set of multiple superset optimal models with respect > > > >to certain objective atoms. > > > > > > > > I attempted to enable cautious reasoning mode by passing the > > > > > > solving > > > > > > >option --enum-mode=cautious to the Clingo Control object used by > > > >Asprin, where a superset preference is specified. However, this did > > > >not produce the desired effect. > > > > > > > > Does anyone know how to enable cautious reasoning mode when using > > > > > > > >Asprin? If this is not possible, are there any alternative > > > > > > approaches > > > > > > >to achieve this? > > > > > > > > Thank you, > > > > Zhiliang > > > > > > _______________________________________________ > > > Potassco-users mailing list > > > Pot...@li... > > > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists. > > > so<https://lists.so/> > > > urceforge.net%2Flists%2Flistinfo%2Fpotassco-users&data=05%7C02%7Cxiangz > > > 6%4 > > > 0cardiff.ac.uk%7Cdb0bfcd8ae3b4e0c2f6308dc8bc2ca4f%7Cbdb74b3095684856bdb > > > f06 > > > 759778fcbc%7C1%7C0%7C638538915931326095%7CUnknown%7CTWFpbGZsb3d8eyJWIjo > > > iMC > > > 4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&s > > > dat > > > a=zj2R2qpv%2FzN4%2FBaTCdLL6hFAAyn4O9t1%2BaDqj29HDj0%3D&reserved=0<https > > > :// lists.sourceforge.net/lists/listinfo/potassco-users> |
|
From: Zhiliang X. <Xi...@ca...> - 2024-06-14 14:35:06
|
Thank you, Roland. Could you please elaborate on what you mean by "However, this relies on a reduction to disjunctive logic program. As such it will come with a hefty performance penalty. "? This seems important to understand, as we tend to encounter scalability issues when the dataset gets larger, even when using Asprin. Best regards, Zhiliang ________________________________ From: Roland Kaminski <kam...@cs...> Sent: 13 June 2024 19:25 To: pot...@li... <pot...@li...> Subject: Re: [Potassco-users] Computing cautious answer set of multiple superset optimal models Hi, approaching this via meta programming would work. The metasp project can easily be adjusted to support a subset maximization certeria. It would even work out of the box with the cautious consequences option. However, this relies on a reduction to disjunctive logic program. As such it will come with a hefty performance penalty. I am not sure it will scale to your use case. -R On Thursday 13 June 2024 20:02:35 CEST Zhiliang Xiang wrote: > Hi Roland, > If I understood correctly, you are suggesting adopting an enumeration scheme > similar to how Clingo enumerates cautious consequences. If that's the case, > I suppose it would likely involve modifications on Asprin's source code on > enumeration, I'm unsure where to begin and how to do this non-intrusively. > Do you have any suggestions in this regard? In fact, one of my colleagues > recommended trying meta-programming [1] for optimisation, then combining it > with cautious reasoning mode. What do you think? Would this be a plausible > approach? Thank you very much, > Zhiliang > > [1] > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.cs.uni-potsdam.de%2Fwv%2Fpublications%2FTEMP_journals%2Ftplp%2FGebserKS11&data=05%7C02%7Cxiangz6%40cardiff.ac.uk%7C3b135f868fb449af6e1708dc8bd6f273%7Cbdb74b3095684856bdbf06759778fcbc%7C1%7C0%7C638539002515068396%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=JD0hdb13V8iFPuNHL%2Ftnn1rKfWToAAUp6Zo1MSw4CrU%3D&reserved=0<https://www.cs.uni-potsdam.de/wv/publications/TEMP_journals/tplp/GebserKS11> > x.pdf ________________________________ > From: Roland Kaminski <kam...@cs...> > Sent: 13 June 2024 17:04 > To: pot...@li... > <pot...@li...>; Javier Romero Davila > <jav...@un...> Subject: Re: [Potassco-users] > [Extern] Computing cautious answer set of multiple superset optimal models > > Maybe you can refine the enumeration scheme? If I am not mistaken, you > can still gather a set of atoms that is not part of the cautious > consequences. This set gets larger with each model found. We can add > constraints pruning all models that are subsets of this set. These > constraints should not interfere with the subset maximization. Maybe > this would help... -R > > On Thu, 2024-06-13 at 15:22 +0000, Zhiliang Xiang wrote: > > Hi Javier, > > Thank you for your reply. > > I forgot to mention that I did try to enumerate all the optimal > > models, but unfortunately, the process did not terminate even after > > days, hitting timeout or memory limits. > > Given this, do you have any other suggestions or know of any other > > techniques to approach this problem? > > Best regards, > > Zhiliang > > From: Javier Romero Davila <jav...@un...> > > Sent: 13 June 2024 05:39 > > To: pot...@li... < > > pot...@li...>; Zhiliang Xiang < > > Xi...@ca...> > > Subject: Re: [Extern][Potassco-users] Computing cautious answer set > > of multiple superset optimal models > > > > External email to Cardiff University - Take care when > > replying/opening attachments or links. > > Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth > > ateb/agor atodiadau neu ddolenni. > > > > > > > > Hi Zhliang, > > > > Asprin does not have an option to compute cautious consequences wrt > > optimal > > models. > > You would have to use asprin to compute all optimal models, > > and then compute the intersection yourself. > > > > Best wishes, > > Javier > > > > On Wed, 12 Jun 2024 17:11:20 +0000 > > > > Zhiliang Xiang <Xi...@ca...> wrote: > > > Dear Community, > > > I am currently working on a project that involves computing the > > > > > >cautious answer set of multiple superset optimal models with respect > > >to certain objective atoms. > > > > > > I attempted to enable cautious reasoning mode by passing the > > > > solving > > > > >option --enum-mode=cautious to the Clingo Control object used by > > >Asprin, where a superset preference is specified. However, this did > > >not produce the desired effect. > > > > > > Does anyone know how to enable cautious reasoning mode when using > > > > > >Asprin? If this is not possible, are there any alternative > > > > approaches > > > > >to achieve this? > > > > > > Thank you, > > > Zhiliang > > > > _______________________________________________ > > Potassco-users mailing list > > Pot...@li... > > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.so<https://lists.so/> > > urceforge.net%2Flists%2Flistinfo%2Fpotassco-users&data=05%7C02%7Cxiangz6%4 > > 0cardiff.ac.uk%7Cdb0bfcd8ae3b4e0c2f6308dc8bc2ca4f%7Cbdb74b3095684856bdbf06 > > 759778fcbc%7C1%7C0%7C638538915931326095%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC > > 4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdat > > a=zj2R2qpv%2FzN4%2FBaTCdLL6hFAAyn4O9t1%2BaDqj29HDj0%3D&reserved=0<https:// > > lists.sourceforge.net/lists/listinfo/potassco-users> |
|
From: Roland K. <kam...@cs...> - 2024-06-13 18:28:53
|
Hi, approaching this via meta programming would work. The metasp project can easily be adjusted to support a subset maximization certeria. It would even work out of the box with the cautious consequences option. However, this relies on a reduction to disjunctive logic program. As such it will come with a hefty performance penalty. I am not sure it will scale to your use case. -R On Thursday 13 June 2024 20:02:35 CEST Zhiliang Xiang wrote: > Hi Roland, > If I understood correctly, you are suggesting adopting an enumeration scheme > similar to how Clingo enumerates cautious consequences. If that's the case, > I suppose it would likely involve modifications on Asprin's source code on > enumeration, I'm unsure where to begin and how to do this non-intrusively. > Do you have any suggestions in this regard? In fact, one of my colleagues > recommended trying meta-programming [1] for optimisation, then combining it > with cautious reasoning mode. What do you think? Would this be a plausible > approach? Thank you very much, > Zhiliang > > [1] > https://www.cs.uni-potsdam.de/wv/publications/TEMP_journals/tplp/GebserKS11 > x.pdf ________________________________ > From: Roland Kaminski <kam...@cs...> > Sent: 13 June 2024 17:04 > To: pot...@li... > <pot...@li...>; Javier Romero Davila > <jav...@un...> Subject: Re: [Potassco-users] > [Extern] Computing cautious answer set of multiple superset optimal models > > Maybe you can refine the enumeration scheme? If I am not mistaken, you > can still gather a set of atoms that is not part of the cautious > consequences. This set gets larger with each model found. We can add > constraints pruning all models that are subsets of this set. These > constraints should not interfere with the subset maximization. Maybe > this would help... -R > > On Thu, 2024-06-13 at 15:22 +0000, Zhiliang Xiang wrote: > > Hi Javier, > > Thank you for your reply. > > I forgot to mention that I did try to enumerate all the optimal > > models, but unfortunately, the process did not terminate even after > > days, hitting timeout or memory limits. > > Given this, do you have any other suggestions or know of any other > > techniques to approach this problem? > > Best regards, > > Zhiliang > > From: Javier Romero Davila <jav...@un...> > > Sent: 13 June 2024 05:39 > > To: pot...@li... < > > pot...@li...>; Zhiliang Xiang < > > Xi...@ca...> > > Subject: Re: [Extern][Potassco-users] Computing cautious answer set > > of multiple superset optimal models > > > > External email to Cardiff University - Take care when > > replying/opening attachments or links. > > Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth > > ateb/agor atodiadau neu ddolenni. > > > > > > > > Hi Zhliang, > > > > Asprin does not have an option to compute cautious consequences wrt > > optimal > > models. > > You would have to use asprin to compute all optimal models, > > and then compute the intersection yourself. > > > > Best wishes, > > Javier > > > > On Wed, 12 Jun 2024 17:11:20 +0000 > > > > Zhiliang Xiang <Xi...@ca...> wrote: > > > Dear Community, > > > I am currently working on a project that involves computing the > > > > > >cautious answer set of multiple superset optimal models with respect > > >to certain objective atoms. > > > > > > I attempted to enable cautious reasoning mode by passing the > > > > solving > > > > >option --enum-mode=cautious to the Clingo Control object used by > > >Asprin, where a superset preference is specified. However, this did > > >not produce the desired effect. > > > > > > Does anyone know how to enable cautious reasoning mode when using > > > > > >Asprin? If this is not possible, are there any alternative > > > > approaches > > > > >to achieve this? > > > > > > Thank you, > > > Zhiliang > > > > _______________________________________________ > > Potassco-users mailing list > > Pot...@li... > > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.so > > urceforge.net%2Flists%2Flistinfo%2Fpotassco-users&data=05%7C02%7Cxiangz6%4 > > 0cardiff.ac.uk%7Cdb0bfcd8ae3b4e0c2f6308dc8bc2ca4f%7Cbdb74b3095684856bdbf06 > > 759778fcbc%7C1%7C0%7C638538915931326095%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC > > 4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdat > > a=zj2R2qpv%2FzN4%2FBaTCdLL6hFAAyn4O9t1%2BaDqj29HDj0%3D&reserved=0<https:// > > lists.sourceforge.net/lists/listinfo/potassco-users> |
|
From: Zhiliang X. <Xi...@ca...> - 2024-06-13 18:03:58
|
Hi Roland, If I understood correctly, you are suggesting adopting an enumeration scheme similar to how Clingo enumerates cautious consequences. If that's the case, I suppose it would likely involve modifications on Asprin's source code on enumeration, I'm unsure where to begin and how to do this non-intrusively. Do you have any suggestions in this regard? In fact, one of my colleagues recommended trying meta-programming [1] for optimisation, then combining it with cautious reasoning mode. What do you think? Would this be a plausible approach? Thank you very much, Zhiliang [1] https://www.cs.uni-potsdam.de/wv/publications/TEMP_journals/tplp/GebserKS11x.pdf ________________________________ From: Roland Kaminski <kam...@cs...> Sent: 13 June 2024 17:04 To: pot...@li... <pot...@li...>; Javier Romero Davila <jav...@un...> Subject: Re: [Potassco-users] [Extern] Computing cautious answer set of multiple superset optimal models Maybe you can refine the enumeration scheme? If I am not mistaken, you can still gather a set of atoms that is not part of the cautious consequences. This set gets larger with each model found. We can add constraints pruning all models that are subsets of this set. These constraints should not interfere with the subset maximization. Maybe this would help... -R On Thu, 2024-06-13 at 15:22 +0000, Zhiliang Xiang wrote: > Hi Javier, > Thank you for your reply. > I forgot to mention that I did try to enumerate all the optimal > models, but unfortunately, the process did not terminate even after > days, hitting timeout or memory limits. > Given this, do you have any other suggestions or know of any other > techniques to approach this problem? > Best regards, > Zhiliang > From: Javier Romero Davila <jav...@un...> > Sent: 13 June 2024 05:39 > To: pot...@li... < > pot...@li...>; Zhiliang Xiang < > Xi...@ca...> > Subject: Re: [Extern][Potassco-users] Computing cautious answer set > of multiple superset optimal models > > External email to Cardiff University - Take care when > replying/opening attachments or links. > Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth > ateb/agor atodiadau neu ddolenni. > > > > Hi Zhliang, > > Asprin does not have an option to compute cautious consequences wrt > optimal > models. > You would have to use asprin to compute all optimal models, > and then compute the intersection yourself. > > Best wishes, > Javier > > On Wed, 12 Jun 2024 17:11:20 +0000 > Zhiliang Xiang <Xi...@ca...> wrote: > > Dear Community, > > I am currently working on a project that involves computing the > >cautious answer set of multiple superset optimal models with respect > >to certain objective atoms. > > I attempted to enable cautious reasoning mode by passing the > solving > >option --enum-mode=cautious to the Clingo Control object used by > >Asprin, where a superset preference is specified. However, this did > >not produce the desired effect. > > Does anyone know how to enable cautious reasoning mode when using > >Asprin? If this is not possible, are there any alternative > approaches > >to achieve this? > > Thank you, > > Zhiliang > > > > > > _______________________________________________ > Potassco-users mailing list > Pot...@li... > https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.sourceforge.net%2Flists%2Flistinfo%2Fpotassco-users&data=05%7C02%7Cxiangz6%40cardiff.ac.uk%7Cdb0bfcd8ae3b4e0c2f6308dc8bc2ca4f%7Cbdb74b3095684856bdbf06759778fcbc%7C1%7C0%7C638538915931326095%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=zj2R2qpv%2FzN4%2FBaTCdLL6hFAAyn4O9t1%2BaDqj29HDj0%3D&reserved=0<https://lists.sourceforge.net/lists/listinfo/potassco-users> |
|
From: Zhiliang X. <Xi...@ca...> - 2024-06-13 18:00:08
|
Hi Carmine, Thank you very much for the reference, this indeed looks really relevant! Best, Zhiliang ________________________________ From: Carmine Dodaro <car...@un...> Sent: 13 June 2024 16:47 To: pot...@li... <pot...@li...> Cc: Javier Romero Davila <jav...@un...> Subject: Re: [Potassco-users] [Extern] Computing cautious answer set of multiple superset optimal models Dear Zhiliang, We recently published a paper about this topic https://www.sciencedirect.com/science/article/pii/S0004370223000772?via%3Dihub. We implemented the algorithms for computing cautious consequences in wasp (https://github.com/alviano/wasp). If you are familiar with the Python API of clingo you might reimplement them. Best, Carmine On 13 Jun 2024, at 17:23, Zhiliang Xiang <Xi...@ca...> wrote: Hi Javier, Thank you for your reply. I forgot to mention that I did try to enumerate all the optimal models, but unfortunately, the process did not terminate even after days, hitting timeout or memory limits. Given this, do you have any other suggestions or know of any other techniques to approach this problem? Best regards, Zhiliang ________________________________ From: Javier Romero Davila <jav...@un...> Sent: 13 June 2024 05:39 To: pot...@li... <pot...@li...>; Zhiliang Xiang <Xi...@ca...> Subject: Re: [Extern][Potassco-users] Computing cautious answer set of multiple superset optimal models External email to Cardiff University - Take care when replying/opening attachments or links. Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth ateb/agor atodiadau neu ddolenni. Hi Zhliang, Asprin does not have an option to compute cautious consequences wrt optimal models. You would have to use asprin to compute all optimal models, and then compute the intersection yourself. Best wishes, Javier On Wed, 12 Jun 2024 17:11:20 +0000 Zhiliang Xiang <Xi...@ca...> wrote: > Dear Community, > I am currently working on a project that involves computing the >cautious answer set of multiple superset optimal models with respect >to certain objective atoms. > I attempted to enable cautious reasoning mode by passing the solving >option --enum-mode=cautious to the Clingo Control object used by >Asprin, where a superset preference is specified. However, this did >not produce the desired effect. > Does anyone know how to enable cautious reasoning mode when using >Asprin? If this is not possible, are there any alternative approaches >to achieve this? > Thank you, > Zhiliang > > _______________________________________________ Potassco-users mailing list Pot...@li... https://lists.sourceforge.net/lists/listinfo/potassco-users ------------------------------------------------------------------------------------------------------------------ Il banner è generato automaticamente dal servizio di posta elettronica dell'Università della Calabria [https://www.unical.it/5x1000]<https://www.unical.it/5x1000> |
|
From: Roland K. <kam...@cs...> - 2024-06-13 16:04:57
|
Maybe you can refine the enumeration scheme? If I am not mistaken, you can still gather a set of atoms that is not part of the cautious consequences. This set gets larger with each model found. We can add constraints pruning all models that are subsets of this set. These constraints should not interfere with the subset maximization. Maybe this would help... -R On Thu, 2024-06-13 at 15:22 +0000, Zhiliang Xiang wrote: > Hi Javier, > Thank you for your reply. > I forgot to mention that I did try to enumerate all the optimal > models, but unfortunately, the process did not terminate even after > days, hitting timeout or memory limits. > Given this, do you have any other suggestions or know of any other > techniques to approach this problem? > Best regards, > Zhiliang > From: Javier Romero Davila <jav...@un...> > Sent: 13 June 2024 05:39 > To: pot...@li... < > pot...@li...>; Zhiliang Xiang < > Xi...@ca...> > Subject: Re: [Extern][Potassco-users] Computing cautious answer set > of multiple superset optimal models > > External email to Cardiff University - Take care when > replying/opening attachments or links. > Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth > ateb/agor atodiadau neu ddolenni. > > > > Hi Zhliang, > > Asprin does not have an option to compute cautious consequences wrt > optimal > models. > You would have to use asprin to compute all optimal models, > and then compute the intersection yourself. > > Best wishes, > Javier > > On Wed, 12 Jun 2024 17:11:20 +0000 > Zhiliang Xiang <Xi...@ca...> wrote: > > Dear Community, > > I am currently working on a project that involves computing the > >cautious answer set of multiple superset optimal models with respect > >to certain objective atoms. > > I attempted to enable cautious reasoning mode by passing the > solving > >option --enum-mode=cautious to the Clingo Control object used by > >Asprin, where a superset preference is specified. However, this did > >not produce the desired effect. > > Does anyone know how to enable cautious reasoning mode when using > >Asprin? If this is not possible, are there any alternative > approaches > >to achieve this? > > Thank you, > > Zhiliang > > > > > > _______________________________________________ > Potassco-users mailing list > Pot...@li... > https://lists.sourceforge.net/lists/listinfo/potassco-users |
|
From: Carmine D. <car...@un...> - 2024-06-13 15:47:24
|
Dear Zhiliang, We recently published a paper about this topic https://www.sciencedirect.com/science/article/pii/S0004370223000772?via%3Dihub. We implemented the algorithms for computing cautious consequences in wasp (https://github.com/alviano/wasp). If you are familiar with the Python API of clingo you might reimplement them. Best, Carmine > On 13 Jun 2024, at 17:23, Zhiliang Xiang <Xi...@ca...> wrote: > > > Hi Javier, > Thank you for your reply. > I forgot to mention that I did try to enumerate all the optimal models, but unfortunately, the process did not terminate even after days, hitting timeout or memory limits. > Given this, do you have any other suggestions or know of any other techniques to approach this problem? > Best regards, > Zhiliang > From: Javier Romero Davila <jav...@un...> > Sent: 13 June 2024 05:39 > To: pot...@li... <pot...@li...>; Zhiliang Xiang <Xi...@ca...> > Subject: Re: [Extern][Potassco-users] Computing cautious answer set of multiple superset optimal models > > External email to Cardiff University - Take care when replying/opening attachments or links. > Nid ebost mewnol o Brifysgol Caerdydd yw hwn - Cymerwch ofal wrth ateb/agor atodiadau neu ddolenni. > > > > Hi Zhliang, > > Asprin does not have an option to compute cautious consequences wrt optimal > models. > You would have to use asprin to compute all optimal models, > and then compute the intersection yourself. > > Best wishes, > Javier > > On Wed, 12 Jun 2024 17:11:20 +0000 > Zhiliang Xiang <Xi...@ca...> wrote: > > Dear Community, > > I am currently working on a project that involves computing the > >cautious answer set of multiple superset optimal models with respect > >to certain objective atoms. > > I attempted to enable cautious reasoning mode by passing the solving > >option --enum-mode=cautious to the Clingo Control object used by > >Asprin, where a superset preference is specified. However, this did > >not produce the desired effect. > > Does anyone know how to enable cautious reasoning mode when using > >Asprin? If this is not possible, are there any alternative approaches > >to achieve this? > > Thank you, > > Zhiliang > > > > > > _______________________________________________ > Potassco-users mailing list > Pot...@li... > https://lists.sourceforge.net/lists/listinfo/potassco-users -- ------------------------------------------------------------------------------------------------------------------ Il banner è generato automaticamente dal servizio di posta elettronica dell'Università della Calabria <https://www.unical.it/5x1000> |
28 messages has been excluded from this view by a project administrator.