Finding All $\epsilon$-Good Arms in Stochastic Bandits

Authors: Blake Mason, Lalit Jain, Ardhendu Tripathy, Robert Nowak

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

Reproducibility Variable Result LLM Response
Research Type Experimental We introduce two algorithms to overcome these and demonstrate their great empirical performance on a large-scale crowd-sourced dataset of 2.2M ratings collected by the New Yorker Caption Contest as well as a dataset testing hundreds of possible cancer drugs.
Researcher Affiliation Academia Blake Mason University of Wisconsin Madison, WI 53706 bmason3@wisc.edu Lalit Jain University of Washington Seattle, WA 98115 lalitj@uw.edu Ardhendu Tripathy University of Wisconsin Madison, WI 53706 astripathy@wisc.edu Robert Nowak University of Wisconsin Madison, WI 53706 rdnowak@wisc.edu
Pseudocode Yes Algorithm 1 (ST)2: Sample the Threshold, Split the Threshold ... Algorithm 4.1: additive FAREAST with γ = 0
Open Source Code Yes Implementations of all algorithms and baselines used in this paper are available on Git Hub.
Open Datasets No The paper mentions using a "large-scale crowd-sourced dataset of 2.2M ratings collected by the New Yorker Caption Contest" and a "dataset [27] of 189 inhibitors". While it cites a paper for the cancer drug dataset, it does not provide a direct link, DOI, or repository access information for either dataset.
Dataset Splits No The paper does not specify explicit training, validation, or test dataset splits (e.g., percentages or sample counts).
Hardware Specification No The paper does not provide specific details about the hardware used for running experiments, such as CPU/GPU models, memory, or cloud instance types.
Software Dependencies No The paper does not list any specific software dependencies with version numbers (e.g., programming languages, libraries, or frameworks).
Experiment Setup Yes In the first example on the left, δ = 0.1, = β = 0.05. ... for δ = 0.01... We set = 0.1 and focus on the multiplicative setting... In this experiment, we use the multiplicative case of ALLwith = 0.8 and δ = 0.001.