Regret of Queueing Bandits

Authors: Subhashini Krishnasamy, Rajat Sen, Ramesh Johari, Sanjay Shakkottai

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

Reproducibility Variable Result LLM Response
Research Type Experimental In this section we present simulation results of various queueing bandit systems with K servers. These results corroborate our theoretical analysis in Sections 4 and 5.
Researcher Affiliation Academia Subhashini Krishnasamy University of Texas at Austin Rajat Sen University of Texas at Austin Ramesh Johari Stanford University Sanjay Shakkottai University of Texas at Austin
Pseudocode Yes Details of the algorithm are given in Algorithm 1 in the Appendix.
Open Source Code No The paper does not provide any concrete statement about the availability of source code for the methodology described, nor does it include a link to a code repository.
Open Datasets No The paper describes a simulated queueing system with generated arrivals and service according to Bernoulli distributions based on specified parameters (K, lambda, mu), rather than using a publicly available dataset.
Dataset Splits No The paper describes a simulated queueing system and evaluates its performance over time 't', but it does not involve traditional dataset splits (e.g., train/validation/test) in its experimental setup.
Hardware Specification No The paper does not provide specific details about the hardware (e.g., CPU, GPU models, or cloud computing instances) used to run the simulations or experiments.
Software Dependencies No The paper describes the Q-Th S algorithm (Algorithm 1) and its theoretical analysis and simulation results, but it does not specify any software dependencies with version numbers (e.g., programming languages, libraries, or frameworks).
Experiment Setup Yes Figure 1 shows the evolution of queue-regret with time in a system with K = 5, = 0.1 and = 0.17. ... At time-slot t, Q-Th S decides to explore with probability min{1, 3K log2 t/t}, otherwise it exploits.