Enhancing Lazy Grounding with Lazy Normalization in Answer-Set Programming
Authors: Jori Bomanson, Tomi Janhunen, Antonius Weinzierl2694-2702
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Benchmark results indicate that lazy normalization can bring up-to exponential gains in space and time as well as enable ASP to be used in new application areas.We therefore evaluate our approach, implemented in the lazy-grounding ASP system Alpha (Weinzierl 2017), on a custom, yet meaningful, set of benchmarks. Since no other lazy-grounding ASP system supports aggregates, we can only compare with ground-and-solve systems and chose Clingo (Gebser et al. 2016) and Dlv2 (Alviano et al. 2017) as representatives. |
| Researcher Affiliation | Academia | 1Department of Computer Science, Aalto University, Finland 2Institute of Logic and Computation, TU Wien, Austria |
| Pseudocode | Yes | Definition 6. The counting grid evaluation program S is: span(R, 1..I 1) idx(R, I). sum(R, 0, 0) idx(R, ). sum(R, I, S) sum(R, I 1, S), span(R, I). sum(R, I, S+1) sum(R, I 1, S), idx(R, I), bound(R, K), S < K. out(R, K) bound(R, K), K S, sum(R, , S). |
| Open Source Code | No | No concrete access to source code for the methodology described in this paper is provided. The paper mentions 'proof-of-concept implementations of lazy normalizations within the Alpha solver' but does not provide a link or statement of availability. |
| Open Datasets | No | The paper evaluates on 'a custom, yet meaningful, set of benchmarks' and mentions 'random instances' or '(random) graph' for its benchmarks, but does not provide concrete access information (link, DOI, citation with authors/year) for any publicly available or open dataset. |
| Dataset Splits | No | The paper evaluates the performance of an ASP solver and does not describe dataset splits (training, validation, test) for reproducibility. |
| Hardware Specification | Yes | each benchmark had 300 seconds and 8GB of memory on a single core of a Linux cluster with Intel Xeon E5-2680 v3 CPUs. |
| Software Dependencies | Yes | The evaluated version of Clingo is 5.2.2 and the version of Dlv2 is 2.0. |
| Experiment Setup | Yes | For each instance we report CPU time averaged over 10 runs of computing the first 10 answer sets. If not otherwise indicated, each benchmark had 300 seconds and 8GB of memory on a single core of a Linux cluster with Intel Xeon E5-2680 v3 CPUs. |