Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1]

Constraining Information Sharing to Improve Cooperative Information Gathering

Authors: Igor Rochlin, David Sarne

JAIR 2015 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Using synthetic environments, we numerically demonstrate that all five methods result in substantial improvement to each of the agents individual expected benefit for a wide range of settings. The results contribute to the advancement of theories of cooperation in MAS.
Researcher Affiliation Academia Igor Rochlin EMAIL School of Computer Science, College of Management, Rishon Le Zion, Israel. David Sarne EMAIL Department of Computer Science, Bar-Ilan University, Ramat-Gan, Israel.
Pseudocode No The paper includes theorems, proofs, and mathematical formulations, but it does not contain any clearly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not contain any explicit statement about releasing source code, nor does it provide links to any code repositories or supplementary materials containing code.
Open Datasets No For simplicity and ease of exposition of the figures, we use f(y) to be the uniform distribution function (between 0 and 1). We stress that even though such a homogeneous setting is standard in costly information gathering literature (Mc Millan & Rothschild, 1994; Lippman & Mc Call, 1976), and common in real-life ISP as argued above, its use in our case is merely for illustration purposes and all the results concerning individual strategies and equilibrium structures that are given in this paper are based on formal theoretical proofs.
Dataset Splits No The paper uses synthetic data generated from a uniform distribution for numerical demonstrations. The concept of training/test/validation splits is not applicable in this context, and no such splits are mentioned.
Hardware Specification No The paper provides theoretical analysis and numerical demonstrations but does not specify any particular hardware (CPU, GPU, memory, etc.) used for running these experiments.
Software Dependencies No The paper discusses mathematical models and presents numerical results; however, it does not mention specific software libraries, programming languages, or tools with version numbers used for these demonstrations.
Experiment Setup Yes The setting used is the homogeneous setting described at the beginning of the section and the value of PISi is the same for all agents (i.e., PISi = PISi). The other model parameters were set to: c = 0.35 and n = 5 (Figure 1(a)) and k = 15 and n = 4 (Figure 1(b)).