High Dimensional Bayesian Optimization with Elastic Gaussian Process

Authors: Santu Rana, Cheng Li, Sunil Gupta, Vu Nguyen, Svetha Venkatesh

ICML 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental experiments clearly demonstrate the utility of the proposed method on both benchmark test functions and real-world case studies. We evaluate our method on three different benchmark test functions and two real-world applications including training cascaded classifiers and for alloy composition optimization.
Researcher Affiliation Academia 1Centre for Pattern Recognition and Data Analytics (PRa DA), Deakin University, Australia.
Pseudocode Yes Algorithm 1 High Dimensional Bayesian Optimization with Elastic Gaussian Process. Algorithm 2 Optimizing acquistion function using EGP.
Open Source Code No The code is available on request.
Open Datasets Yes Training cascade classifier on three real datasets from UCI repository (Blake and Merz, 1998): Ionosphere, German and IJCNN1.
Dataset Splits No The paper does not explicitly provide training/validation/test dataset splits (e.g., percentages, counts, or specific cross-validation schemes) needed to reproduce the data partitioning.
Hardware Specification Yes The computer used is a Xeon Quad-core PC running at 2.6 GHz, with 16 GB of RAM.
Software Dependencies No The paper mentions software like NLopt and Matlab but does not provide specific version numbers for these or other key libraries/dependencies.
Experiment Setup Yes We use the target length-scale lτ = 0.1, lmax = d and lmin = 10 5. The number of initial observations are set at d + 1. The optimization time for all these high-dimensional optimization problem is set as Topt = 0.1 d sec.