Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
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.
Online Learning of Delayed Choices
Authors: Recep Yusuf Bekci
NeurIPS 2024 | Venue PDF | LLM Run Details | Input Tokens: 26,068 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 4,106 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct experiments that confirm the effectiveness of our algorithm. 7 Experiments: We conducted two sets of experiments to evaluate the performance of our algorithms. |
| Researcher Affiliation | Academia | Recep Yusuf Bekci University of Waterloo Waterloo, Canada EMAIL |
| Pseudocode | Yes | Algorithm 1 Delayed MNL Bandit (DEMBA) |
| Open Source Code | Yes | The necessary information is provided in the Experiments section and in the appendix as well as the scripts for experiments are provided. |
| Open Datasets | No | The paper uses a synthetic dataset generated based on defined parameters (N=10, K=4, pi=1) and attraction parameters, rather than an existing public dataset. No access information is provided for this generated data. |
| Dataset Splits | No | The paper uses a simulation-based approach with synthetic data and measures cumulative regret over rounds, thus standard training, validation, and test dataset splits are not applicable or mentioned. |
| Hardware Specification | Yes | The simulations were conducted on a server equipped with 4 Intel Xeon 6248 2.5GHz CPUs and 377 GB of RAM, running Cent OS 7. |
| Software Dependencies | Yes | The simulation code was developed in Python version 3.9.6. |
| Experiment Setup | Yes | We used N = 10, K = 4 and pi = 1 for all i {1, . . . , N}. The attraction parameters were set as: vi = 0.25 + ฯต if i {1, 2, 9, 10} 0.25 otherwise, where ฯต represents the contrast between products. We used geometric delays with E[ds] = 100 and ยต = 100 for the first experiment and E[ds] = 100 for the second experiment. |