Top-k Ranking Bayesian Optimization

Authors: Quoc Phong Nguyen, Sebastian Tay, Bryan Kian Hsiang Low, Patrick Jaillet9135-9143

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We empirically evaluate the performance of MPES using several synthetic benchmark functions, CIFAR-10 dataset, and SUSHI preference dataset.
Researcher Affiliation Academia 1Dept. of Computer Science, National University of Singapore, Republic of Singapore 2Dept. of Electrical Engineering and Computer Science, MIT, USA
Pseudocode No The paper describes procedures and methods but does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code Yes The code is available at https://github.com/sebtsh/Top-k-Ranking-Bayesian Optimization.
Open Datasets Yes We empirically evaluate the performance of MPES using several synthetic benchmark functions, CIFAR-10 dataset, and SUSHI preference dataset.
Dataset Splits No The paper mentions 'initial observations' for BO algorithms but does not specify standard train/validation/test dataset splits with percentages or counts for reproducibility.
Hardware Specification No The paper does not provide any specific hardware details such as CPU/GPU models, memory, or cloud instance types used for running the experiments.
Software Dependencies No The paper does not list specific software dependencies with version numbers (e.g., Python, PyTorch, or other library versions).
Experiment Setup Yes To evaluate MPES, we set |X | to 20 and the number of samples is n = 1000. The numbers of initial observations provided to the BO algorithms are 5, 6, and 12 for experiments with the Forrester, SHC, and Hartmann functions, respectively. Six initial observations are provided to the BO algorithms [for CIFAR-10]. In these experiments, there are 10 initial observations provided to the BO algorithms [for SUSHI].