Cooperative Multi-player Bandit Optimization

Authors: Ilai Bistritz, Nicholas Bambos

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

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the numerical behavior of Algorithm 1 using a congestion game with N = 8 players and R = 4 resources. ... Figure 2: Congestion game with N = 8 players and R = 4 resources (a) Typical realization (b) Performance over 1000 Realizations
Researcher Affiliation Academia Ilai Bistritz, Nicholas Bambos Stanford University {bistritz,bambos}@stanford.edu
Pseudocode Yes Algorithm 1 Cooperative Multi-player Bandit Optimization
Open Source Code No The paper does not provide any statement or link indicating the release of source code for the described methodology.
Open Datasets No The paper describes a 'Numerical Example' using a 'congestion game' simulation environment, not a publicly available dataset with concrete access information.
Dataset Splits No The paper describes a numerical simulation but does not provide specific details on dataset splits (e.g., train/validation/test percentages or counts).
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments, only describing a 'Numerical Example' simulation.
Software Dependencies No The paper does not mention any specific software dependencies with version numbers needed to replicate the experiments.
Experiment Setup Yes We used the step size sequence η (t) = 0.2 t0.9 and the sampling radius sequence δ (t) = 0.2 t0.1 . The memory was M = 6. Players broadcasted one reward value per turn, uniformly at random.