Estimating High-dimensional Non-Gaussian Multiple Index Models via Stein’s Lemma
Authors: Zhuoran Yang, Krishnakumar Balasubramanian, Zhaoran Wang, Han Liu
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We supplement our theoretical findings with simulations. ... In this section, we evaluate the finite-sample error of the proposed estimators on simulated data. |
| Researcher Affiliation | Collaboration | Princeton University, email: {zy6, kb18}@princeton.edu Tencent AI Lab & Northwestern University, email: {zhaoranwang, hanliu.cmu}@gmail.com |
| Pseudocode | No | The paper describes mathematical formulations and algorithmic approaches (e.g., ADMM) but does not include a formally structured pseudocode or algorithm block. |
| Open Source Code | No | The paper does not provide an explicit statement about open-sourcing the code for the methodology or a link to a code repository. |
| Open Datasets | No | The paper states that simulated data from specific distributions (Gamma, Rayleigh) were used, but there is no mention of these simulated datasets or the generation scripts being publicly available or linked. |
| Dataset Splits | No | The paper mentions running '100 independent trials for each n' on simulated data, but it does not specify explicit train/validation/test dataset splits (e.g., percentages or sample counts) typically used for model validation. |
| Hardware Specification | No | The paper does not specify any particular hardware (e.g., GPU/CPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions using the 'alternating direction method of multipliers (ADMM) algorithm' but does not specify any software libraries or frameworks with version numbers used for implementation. |
| Experiment Setup | Yes | Throughout the experiment we vary n and fix d = 500 and s = 5. Also, the support of β is chosen uniformly at random from all the possible subsets of [d] with cardinality s . For each j 2 supp(β ), we set β s γj, where γj s are i.i.d. Rademacher random variables. Furthermore, we fix the regularization parameter λ = 4 log d/n and threshold parameter = 20. In addition, we adopt the cosine distance cos \(bβ, β ) = 1 |hbβ, β i|, to measure the estimation error. |