Multi-Agent Planning with Baseline Regret Minimization

Authors: Feng Wu, Shlomo Zilberstein, Xiaoping Chen

IJCAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on common benchmark problems confirm the benefits of the algorithm compared with the state-of-the-art approaches.
Researcher Affiliation Academia School of Computer Science and Technology, University of Science and Technology of China, CHN College of Information and Computer Sciences, University of Massachusetts Amherst, USA
Pseudocode Yes Algorithm 1: Iterative Belief Generation
Open Source Code No No statement about releasing open-source code for the described methodology is found.
Open Datasets Yes We empirically evaluated our algorithm on four common benchmark problems1 widely used in the DEC-POMDP literature: Broadcast Channel, Recycling Robots, Cooperative Box Pushing, and Meeting in a 3 3 Grid. 1http://masplan.org/problem_domains
Dataset Splits No The paper discusses running experiments on benchmark problems and problem instances, but does not specify training, validation, or test dataset splits.
Hardware Specification Yes Our algorithm (i.e., IBG-DP) was implemented in Java 1.8 and ran on a machine with 3.5GHz Intel Core i7 CPU and 8GB of RAM.
Software Dependencies Yes Our algorithm (i.e., IBG-DP) was implemented in Java 1.8 and ran on a machine with 3.5GHz Intel Core i7 CPU and 8GB of RAM. The MILP and LP were solved by IBM CPLEX 12.61.
Experiment Setup Yes The baseline policies that we used here are random policy trees with max Tree = 3.