Exploiting Support Sets for Answer Set Programs with External Evaluations
Authors: Thomas Eiter, Michael Fink, Christoph Redl, Daria Stepanova
AAAI 2014 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We compare an implementation of the novel against the traditional approach. The results show its effectiveness, with significant gains (up to two orders of magnitude; Section 6). ... The average runtimes on 100 randomly generated instances for graphs with n nodes are shown in Table 1, where Sup is traditional and +Sup support set evaluation; numbers in parentheses are timeouts. |
| Researcher Affiliation | Academia | Thomas Eiter and Michael Fink and Christoph Redl and Daria Stepanova Institute of Information Systems, Vienna University of Technology Favoritenstraße 9-11, A-1040 Vienna, Austria {eiter,fink,redl,dasha}@kr.tuwien.ac.at |
| Pseudocode | No | The paper describes algorithms and formalizations in prose and with mathematical notation but does not include any explicitly labeled pseudocode blocks or algorithms. |
| Open Source Code | No | The paper mentions "Program instances used in the experiments are available at http://www.kr.tuwien.ac.at/research/projects/hexhex/supportsets/". This link provides instances (data/benchmarks) rather than the source code for the methodology itself. There is no explicit statement about releasing the source code of their method. |
| Open Datasets | Yes | Program instances used in the experiments are available at http://www.kr.tuwien.ac.at/research/projects/hexhex/supportsets/. ... Finally, we consider default reasoning over the famous LUBM ontology3 in its DL-Lite A form. |
| Dataset Splits | No | The paper describes randomly generated instances and reports average runtimes, but it does not specify explicit training, validation, or test splits by percentages or counts. It states, for example, "The average runtimes on 100 randomly generated instances for graphs with n nodes". |
| Hardware Specification | Yes | They were run on a Linux server with two 12-core AMD 6176 SE CPUs/128GB RAM using a timeout of 300 secs per run. |
| Software Dependencies | Yes | Both the traditional and the new algorithm are implemented in the DLVHEX system version 2.3.0, where a command-line switch allows to select the algorithm. The system is based on GRINGO and CLASP for either selection. |
| Experiment Setup | Yes | They were run on a Linux server with two 12-core AMD 6176 SE CPUs/128GB RAM using a timeout of 300 secs per run. ... The average runtimes on 100 randomly generated instances for graphs with n nodes are shown in Table 1... Instances of size n have n persons, n+2 cabinets, n+1 rooms, and 2n objects randomly assigned to the persons; 2n 2 objects are already stored. ... we fixed the ABox A of the ontology O to 50 customers, 20 drivers (among them 4 driving electro-cars), and 5 regions; every driver works in 2-4 regions. |