Optimal Anytime Coalition Structure Generation Utilizing Compact Solution Space Representation
Authors: Redha Taguelmimt, Samir Aknine, Djamila Boukredera, Narayan Changder, Tuomas Sandholm
IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our results show that the proposed algorithm is superior to prior state-of-the-art methods in generating optimal coalition structures under several value distributions. 5 Experiments The main goals of our experiments are to investigate how different input sizes and value distributions affect our search method, and how our algorithm compares to the prior state of the art. |
| Researcher Affiliation | Collaboration | 1Univ Lyon, UCBL, CNRS, INSA Lyon, Centrale Lyon, Univ Lyon 2, LIRIS, UMR5205, Lyon, France 2Laboratory of Applied Mathematics, Faculty of Exact Sciences, University of Bejaia, Bejaia, Algeria 3TCG Centres for Research and Education in Science and Technology, Kolkata, India 4Computer Science Department, Carnegie Mellon University, Pittsburgh, USA 5Strategic Machine, Inc. 6Strategy Robot, Inc. 7Optimized Markets, Inc. |
| Pseudocode | Yes | The pseudocode of the algorithm used by the preprocessor is in the appendix. Now, the search process in RDP is carried out on the coalitions of sizes that belong to the optimal set of sizes S = {s1, s2, ..., sk} obtained by the preprocessor. The pseudocode of RDP is in the appendix. |
| Open Source Code | No | The paper does not provide an explicit statement about releasing its source code or a link to a repository for the described methodology. |
| Open Datasets | Yes | In accordance with established practices in coalition structure generation, we benchmark on the following coalition value distributions: Normal [Rahwan et al., 2007], Modified Normal [Rahwan et al., 2012], Uniform [Larson and Sandholm, 2000], Modified Uniform [Service and Adams, 2010], Beta, Exponential, Gamma [Michalak et al., 2016], Pascal and Weibull [Changder et al., 2020]. |
| Dataset Splits | No | The paper does not specify training, validation, or test dataset splits. It mentions using "50 generated problem instances per value distribution" but not how these instances are split for model development or evaluation phases. |
| Hardware Specification | Yes | The experiments were conducted on an Intel Xeon 2.30GHz E5-2650 CPU with 256GB of RAM. |
| Software Dependencies | No | The paper states that ELIXIR was implemented "in Java" but does not specify a version number for Java or any other software dependencies. |
| Experiment Setup | Yes | ELIXIR uses a hyperparameter for the number of variables that it considers in CSSA. This hyperparameter is optimized using 9000 problem instances by varying the number of variables. Our hyperparameter search and the final hyperparameter are presented in the appendix. |