Preselection Bandits

Authors: Viktor Bengs, Eyke Hüllermeier

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

Reproducibility Variable Result LLM Response
Research Type Experimental We devote Section 6 to a simulation study demonstrating the usefulness and effi ciency of our algorithms. In this section, we investigate the performance of TRCB (Algorithm 1) as well as CBR (Algorithm 2) on synthetic data for some specific scenarios, while providing further scenarios in the supplementary material.
Researcher Affiliation Academia 1Heinz Nixdorf Institute and Department of Computer Science, Paderborn University, Germany.
Pseudocode Yes Algorithm 1 TRCB algorithm; Algorithm 2 CBR-algorithm
Open Source Code No The paper does not provide any statement or link indicating that the source code for its methodology is openly available.
Open Datasets No We investigate the performance of TRCB (Algorithm 1) as well as CBR (Algorithm 2) on synthetic data for some specific scenarios... The score parameters θ = (θi)i [n] are drawn uniformly at random from the n-simplex...
Dataset Splits No The paper states it uses 'randomly generated restricted Pre-Bandit instances' and 'randomly generated flexible Pre-Bandit instances' but does not specify any dataset splits (e.g., training, validation, test percentages or counts).
Hardware Specification No The authors gratefully acknowledge financial support by the Germany Research Foundation (DFG). Moreover, the authors would like to thank the Paderborn Center for Parallel Computation (PC2) for the use of the OCu LUS cluster.
Software Dependencies No The paper does not specify any software names with version numbers used for implementation or dependencies.
Experiment Setup Yes We consider the case n = 10, l = 3, and time horizons T ∈ {i · 2000}i=1. The degree of preciseness is γ = 1 throughout, and the score parameters θ = (θi)i∈[n] are drawn uniformly at random from the n-simplex... In the right picture of Figure 1, the results are displayed for the CBR resp. DTS algorithm on 1000 repetitions, respectively, with n ∈ {5, 10, 15}, T ∈ {i · 2000}5 i=1 , and σ(x) = (1 − x)1[0,)(x).