Agile Planning for Real-World Disaster Response

Authors: Feng Wu, Sarvapali D. Ramchurn, Wenchao Jiang, Jeol E. Fischer, Tom Rodden, Nicholas R. Jennings

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically evaluate our algorithm and show that it outperforms current benchmarks. Our algorithm is also shown to perform better in pilot studies with real humans.
Researcher Affiliation Academia Feng Wu* Sarvapali D. Ramchurn Wenchao Jiang Jeol E. Fischer Tom Rodden Nicholas R. Jennings *Computer Science and Technology, University of Science and Technology of China, Hefei, China Electronics and Computer Science, University of Southampton, Southampton, UK Mixed Reality Lab, University of Nottingham, Nottingham, UK
Pseudocode Yes Algorithm 1: Two-Pass UCT Planning
Open Source Code No A video of our pilot runs can be viewed at: http://bit.ly/1eb NYty. This link leads to a video demonstration, not source code for the methodology.
Open Datasets No We built an MMDP simulator for this scenario. The paper describes a custom simulation environment but does not mention making the dataset or simulation data publicly available.
Dataset Splits No The paper mentions running simulations and pilot studies but does not specify details on train/validation/test splits, percentages, or explicit sample counts for dataset partitioning.
Hardware Specification No The paper does not provide specific hardware details (e.g., CPU, GPU models, or memory specifications) used for running the experiments.
Software Dependencies No The paper mentions building an MMDP simulator and using UCT, but it does not specify any software dependencies with version numbers (e.g., Python 3.x, PyTorch 1.x).
Experiment Setup Yes In the experiments, we initialize the rejection model by randomly generating the preference of each FR and set the discount factor γ = 0.95, the rejection limit k = 3, and the rejection cost C = 1.0.