Active Set Ordering
Authors: Quoc Phong Nguyen, Sunil Gupta, Svetha Venkatesh, Bryan Kian Hsiang Low, Patrick Jaillet
NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We empirically validate the performance of our proposed solution using various synthetic functions and real-world datasets. In Sec. 5, we empirically validate the performance of our solution using several synthetic functions and real-world datasets. |
| Researcher Affiliation | Academia | 1Applied Artificial Intelligence Institute, Deakin University, Australia 2School of Computing, National University of Singapore, Singapore 3LIDS and EECS, Massachusetts Institute of Technology, USA |
| Pseudocode | Yes | Algorithm 1 Mean Prediction (MP) for Active Set Ordering. Algorithm 2 Mean Prediction (MP) for Active Multiple Set Ordering. |
| Open Source Code | Yes | We have provided the source code, datasets, and scripts to reproduce the experiment results. |
| Open Datasets | Yes | To perform experiments with the NO3 dataset from Lake Zurich (available at https://wldb.ilec.or.jp/Lake/EUR-06/datalist), we standardize the NO3 measurements. To perform experiments with the humidity dataset, we extract the humidity measurements at different locations with the same mote id of 31167 from the Intel Lab data (available at https://db.csail.mit.edu/labdata/labdata.html). |
| Dataset Splits | No | The paper mentions data discretization (e.g., 'discretized into a set of 100 points') and generating observations with Gaussian noise, but does not explicitly specify train/validation/test splits with percentages, counts, or predefined standard splits for reproducibility. |
| Hardware Specification | Yes | All experiments were conducted on a computer equipped with an AMD Ryzen 7 6800HS processor and 16GB of RAM. |
| Software Dependencies | No | The paper does not specify any software dependencies with version numbers (e.g., 'Python 3.8', 'PyTorch 1.9'). |
| Experiment Setup | Yes | The noise is chosen with σn = 0.1. The input domain is discretized into a set of 100 points, whereas for the Hartmann-6D function, it is discretized into a set of 1000 points. The experiments are repeated 15 times to account for the randomness in the generation of the observations. |