Generalized Linear Model Regression under Distance-to-set Penalties
Authors: Jason Xu, Eric Chi, Kenneth Lange
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Applications to shape constraints, sparse regression, and rank-restricted matrix regression on synthetic and real data showcase strong empirical performance, even under non-convex constraints. ... We first compare the performance of our distance penalization method to leading shrinkage methods in sparse regression. Our simulations involve a sparse length n = 2000 coefficient vector β with 10 nonzero entries. ... We consider two real datasets. |
| Researcher Affiliation | Academia | Jason Xu University of California, Los Angeles jqxu@ucla.edu Eric C. Chi North Carolina State University eric_chi@ncsu.edu Kenneth Lange University of California, Los Angeles klange@ucla.edu |
| Pseudocode | Yes | Algorithm 1 MM algorithm to solve distance-penalized objective (4) |
| Open Source Code | No | The paper does not provide an unambiguous statement or link indicating that the source code for the methodology described is publicly available. |
| Open Datasets | Yes | We apply our method to count data of global temperature anomalies relative to the 1961-1990 average, collected by the Climate Research Unit [17]. ... We next focus on rank constrained matrix regression for electroencephalography (EEG) data, collected by [35] to study the association between alcoholism and voltage patterns over times and channels. |
| Dataset Splits | No | The paper describes the generation of synthetic data and the characteristics of real datasets, but it does not specify explicit training, validation, or test dataset splits (e.g., percentages, sample counts, or clear cross-validation setup) needed for reproduction. |
| Hardware Specification | No | The paper states that the algorithms "require several seconds of compute time on a standard laptop machine" but does not provide specific hardware details such as GPU/CPU models, memory, or other specifications used for running the experiments. |
| Software Dependencies | No | The paper mentions different algorithms and methods (e.g., LQA, LLA, coordinate-descent implementations) but does not provide specific software names with version numbers for any libraries, frameworks, or tools used in the implementation. |
| Experiment Setup | No | The paper describes aspects of data generation for simulations and general problem settings, but it does not provide specific hyperparameter values (e.g., learning rates, specific penalty weights like `vi`, epochs) or detailed training configurations for the MM algorithm beyond general principles like Armijo backtracking. |