Stochastic Market Games

Authors: Kyrill Schmid, Lenz Belzner, Robert Müller, Johannes Tochtermann, Claudia Linnhoff-Popien

IJCAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically find that the presence of markets can improve both the overall result and agent individual returns via their trading activities.
Researcher Affiliation Academia 1LMU Munich 2Technische Hochschule Ingolstadt
Pseudocode Yes pseudocode given in Algorithm 1
Open Source Code No The paper does not provide any statement or link indicating that the source code for the described methodology is publicly available.
Open Datasets No The paper introduces custom domains called 'Smartfactory' and 'Refinery' for evaluation and does not provide access information or citations for publicly available datasets.
Dataset Splits No The paper describes training processes and evaluation runs but does not specify train/validation/test dataset splits, percentages, or sample counts.
Hardware Specification No The paper does not provide any specific details about the hardware used for running the experiments, such as GPU or CPU models.
Software Dependencies No The paper refers to various algorithms and models (DQN, PPO, MADDPG) but does not specify any software dependencies with version numbers.
Experiment Setup Yes For the experiment we set R1 = 0.375 and R2 = 2.625...two independent Q-learners were trained over 25 independent runs for each level of conflict α, where each run consisted of 4000 consecutive steps. ... here we used p = 1.9. ... In all runs we set rhigh = 5, rlow = 1 tinactive = 8 in the Smartfactory and rhigh = 5, rlow = 0.02 in the Refinery. ... All settings were tested with 4, 8 and 16 agents...