Multi-Variable Agents Decomposition for DCOPs
Authors: Ferdinando Fioretto, William Yeoh, Enrico Pontelli
AAAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our MVA decomposition with three global DCOP algorithms (AFB, DPOP, and D-Gibbs) and two local centralized solvers (DFBn B and Gibbs) implemented on CPUs and GPUs. |
| Researcher Affiliation | Academia | Ferdinando Fioretto Department of Computer Science New Mexico State University Las Cruces, NM 88003, U.S. ffiorett@cs.nmsu.edu |
| Pseudocode | Yes | The procedure GPU-GIBBS is the core of the local sampling algorithm and corresponds to one of the R independent sampling processes executed in parallel. It executes T sampling trials for the subset of non-interface local variables Li \ Bi of agent ai. Its pseudocode, executed by each thread in a group, is shown in lines 1-17. |
| Open Source Code | No | The paper does not explicitly state that the source code for the described methodology is publicly available, nor does it provide a link to a code repository. |
| Open Datasets | No | The paper describes generating its own 'random graph' and 'radar coordination' instances but does not provide concrete access information (link, DOI, repository, or citation) for these datasets to be publicly accessed or reproduced. |
| Dataset Splits | No | The paper does not provide specific dataset split information (e.g., exact percentages, sample counts, or cross-validation setup) for reproducibility. |
| Hardware Specification | Yes | All experiments are performed on an Intel i7 Quadcore 3.4GHz machine with 16GB of RAM. The GPU solvers are run on an Ge Force GTX TITAN with 14 multiprocessors, 2688 cores, and a clock rate of 837MHz. |
| Software Dependencies | No | The paper mentions various algorithms and solvers (e.g., DFBn B, Gibbs, AFB, DPOP, D-Gibbs) but does not specify any software dependencies with version numbers for reproducibility. |
| Experiment Setup | Yes | We impose a timeout (t.o.) of 600sec of simulated time and a memory limit of 2GB. Results report the average over 50 runs, and are statistically significant with p-values < 0.001. We set |Di| = 4, pl 1 = 0.6, and the constraint tightness p2 = 0.4. |