Nonparametric Inference under B-bits Quantization
Authors: Kexuan Li, Ruiqi Liu, Ganggang Xu, Zuofeng Shang
JMLR 2024 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive simulation studies together with a real-data analysis are used to demonstrate the validity and effectiveness of the proposed tests. |
| Researcher Affiliation | Collaboration | Kexuan Li EMAIL Global Biometrics and Data Sciences Bristol Myers Squibb Princeton Pike, NJ 08648, USA; Ruiqi Liu EMAIL Department of Mathematics and Statistics Texas Tech University Lubbock, TX 79409, USA; Ganggang Xu EMAIL Department of Management Science University of Miami Coral Gables, FL 33146, USA; Zuofeng Shang EMAIL Department of Mathematical Sciences New Jersey Institute of Technology Newark, NJ 07102, USA |
| Pseudocode | Yes | Algorithm 1: Two-Stage Quantization; Algorithm 2: Quantization Estimation of Variance |
| Open Source Code | No | The paper does not contain any explicit statements about releasing source code for the methodology, nor does it provide a link to a code repository. |
| Open Datasets | Yes | In this section, we apply the proposed methods to the Combined Cycle Power Plant Data (Kaya et al., 2012; T ufekci, 2014), which can be downloaded at http://https://archive.ics.uci.edu/ml/datasets/Combined+Cycle+Power+Plant. |
| Dataset Splits | No | The paper mentions generating data for simulations with various sample sizes (e.g., n = 1000, 2000, 3000, 5000, 10000) and using a real dataset of n = 9568 observations. However, it does not specify any training, validation, or test splits for these datasets. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used to run the simulation studies or the real-data analysis experiments. |
| Software Dependencies | No | The paper does not list specific software dependencies with version numbers. |
| Experiment Setup | Yes | For all simulation studies, we consider the uniform quantization scheme outlined in Section 2.2. Specifically, for the data quantization step, for a given bits budget B, we choose c, k following the approach suggested in Section 2.5 with a Tn/σ2 = 2.5 log(n). For each simulation, the quantization ranges t1, tk 1 are defined as t1 = µ0 p 2.5σ2 log(n), tk 1 = µ0+ p 2.5σ2 log(n), where µ0 = R 1 0 g0(x)dx with g0( ) being the regression function in model (1). The target significance level was chosen as α = 0.1. ... The tuning parameter λ was set as λ = bλGCV/ log(c) with bλGCV being picked by GCV. |