Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Adaptive Monte Carlo via Bandit Allocation
Authors: James Neufeld, Andras Gyorgy, Csaba Szepesvari, Dale Schuurmans
ICML 2014 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct experimental investigations in a number of scenarios to better understand the effectiveness of multi-armed bandit algorithms for adaptive Monte Carlo estimation. |
| Researcher Affiliation | Academia | James Neufeld EMAIL Andr as Gy orgy EMAIL Dale Schuurmans EMAIL Csaba Szepesv ari EMAIL Department of Computing Science, University of Alberta, Edmonton, AB, Canada T6G 2E8 |
| Pseudocode | Yes | Algorithm 1 UCB1 (Auer et al., 2002a) ... Algorithm 2 Thompson Sampling (Agrawal & Goyal, 2012) |
| Open Source Code | No | The paper does not provide any statement or link regarding the release of open-source code for the methodology described. |
| Open Datasets | Yes | on different sized subsets of the 8-dimensional Pima Indian diabetes UCI data set (Bache & Lichman, 2013). |
| Dataset Splits | No | The paper mentions using 'different sized subsets' of a dataset but does not provide specific train/validation/test split percentages, sample counts, or refer to predefined splits. |
| Hardware Specification | No | The paper mentions using 'elapsed CPU-time' and 'JAVA VM' for cost accounting, but does not provide specific hardware details like CPU/GPU models, memory, or processor types used for running the experiments. |
| Software Dependencies | No | The paper mentions 'JAVA VM' in the context of CPU-time measurement but does not list specific software dependencies with version numbers needed to replicate the experiment. |
| Experiment Setup | Yes | for UCB-V we used the same settings as (Audibert et al., 2009), and for TS we used the uniform Beta prior, i.e., α0 = 1 and β0 = 1. ... we approximated the option prices under the same parameter settings as (Douc et al., 2007), namely, ν = 0.016, κ = 0.2, r0 = 0.08, T = 1, M = 1000, σ = 0.02, and n = 100, for different strike prices K = {0.06, 0.07, 0.08}. ... three AIS estimators that differ only in the number of annealing steps they use; namely, 400, 2000, and 8000 steps. In each case, we fix the annealing schedule using the power of 4 heuristic suggested by (Kuss & Rasmussen, 2005), with a single slice sampling MCMC transition used at each step. |