Co-Optimizating Multi-Agent Placement with Task Assignment and Scheduling

Authors: Chongjie Zhang, Julie A. Shah

IJCAI 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Empirical results demonstrate that this multi-abstraction approach significantly outperforms a conventional hill climbing algorithm and an approximate mixedinteger linear programming approach.
Researcher Affiliation Academia Chongjie Zhang and Julie A. Shah Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology Cambridge, MA 02139 {chongjie,julie a shah}@csail.mit.edu
Pseudocode Yes Algorithm 1 Hill Climbing Algorithm
Open Source Code No The paper does not provide an explicit statement or link for open-source code for the methodology described. It mentions using external tools like Gurobi and Tercio, but not its own implementation.
Open Datasets No The paper states: 'We generated multi-agent task assignment and scheduling problems that simulate multi-agent construction of a cellular assembly line in a manufacturing domain, such as airplane production.' While it mentions a 'task generator similar to that used in [Gombolay et al., 2013]', it does not provide access information (link, DOI, specific citation for *this* generated dataset) for the specific datasets used in *their* experiments.
Dataset Splits No The paper mentions running experiments on '10 scenarios, each with 10 different problems, for a total of 100 problems', but does not specify any training, validation, or test dataset splits within these problems.
Hardware Specification No The paper does not mention any specific hardware used for running the experiments (e.g., CPU, GPU models, memory).
Software Dependencies No The paper mentions 'existing optimizers [Gurobi Optimization, 2015]' and 'Tercio algorithm [Gombolay et al., 2013]'. While these are specific tools/algorithms, explicit version numbers (e.g., Gurobi 7.0) are not provided, only the year of the referenced publication for Gurobi. No other software with version numbers is listed.
Experiment Setup No The paper describes how the synthetic problems were generated (e.g., task time distributions, temporal constraint proportions), but does not provide specific experimental setup details such as hyperparameters, learning rates, batch sizes, or optimizer settings for any models trained or optimized within the MASA framework.