Path Planning Problems with Side Observations—When Colonels Play Hide-and-Seek

Authors: Dong Quan Vu, Patrick Loiseau, Alonso Silva, Long Tran-Thanh2252-2259

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We also conducted several numerical experiments that compare the running time and the actual expected regret of EXP3-OE and FPL-IX in CB and HS games. The numerical results are consistent with theoretical results in this work. Our code for these experiments can be found at https://github.com/dongquan11/CB-HS.SOPPP.
Researcher Affiliation Collaboration 1Nokia Bell Labs France, AAAID Department, 2Univ. Grenoble Alpes, Inria, CNRS, Grenoble INP, LIG & MPI-SWS, 3Safran Tech, Signal and Information Technologies, 4Univ. of Southampton, School of Electronics and Computer Science
Pseudocode Yes Algorithm 1 EXP3-OE Algorithm for SOPPP.
Open Source Code Yes Our code for these experiments can be found at https://github.com/dongquan11/CB-HS.SOPPP.
Open Datasets No The paper describes online learning in adversarial games where losses are chosen at each stage, rather than using a static, publicly available dataset with a training split.
Dataset Splits No The paper focuses on online learning in an adversarial setting and does not mention a traditional validation split for dataset-based experiments.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., "Python 3.8", "PyTorch 1.9").
Experiment Setup No The paper defines parameters like T, η, and β for the EXP3-OE algorithm, but it does not provide specific experimental setup details such as hyperparameters (e.g., learning rate, batch size, epochs), optimizers, or other system-level training settings common in empirical machine learning papers.