Sign-Full Random Projections
Authors: Ping Li4205-4212
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | A Simulation Study We provide a simulation study to verify the theoretical properties of the four estimators for sign-full random projections: ˆρg, ˆρg,n, ˆρs, ˆρs,n, as well as ˆρ1 for sign-sign projections. An Experimental Study To further verify the theoretical results, we conduct an experimental study on the ranking task for near-neighbor search on 4 public datasets (see Table 1 and Figure 5). |
| Researcher Affiliation | Industry | Ping Li Cognitive Computing Lab (CCL) Baidu Research USA Bellevue, WA 98004, USA pingli98@gmail.com |
| Pseudocode | No | The paper includes mathematical formulas and derivations but does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link indicating the release of open-source code for the described methodology. |
| Open Datasets | Yes | We conduct an experimental study on the ranking task for near-neighbor search on 4 public datasets (see Table 1 and Figure 5). These four datasets are downloaded from either the UCI repository or the LIBSVM website. Table 1: Information about the datasets Dataset # Train # Query # Dim MNIST 10,000 10,000 780 RCV1 10,000 10,000 47,236 Youtube Audio 10,000 11,930 2,000 Youtube Description 10,000 11,743 12,183,626 |
| Dataset Splits | No | The paper mentions 'training samples' and a 'query set' but does not provide specific percentages or counts for training, validation, and test splits, nor does it refer to standard predefined splits with citations for reproducibility. |
| Hardware Specification | No | The paper does not provide any specific details regarding the hardware used for running the experiments or simulations. |
| Software Dependencies | No | The paper does not provide specific software dependencies with version numbers for reproducibility. |
| Experiment Setup | Yes | Figure 6 presents the results for the RCV1 datasets, for ρ0 {0.9, 0.8, 0.6}, and for k {50, 100}. |