Implicit Regularization Leads to Benign Overfitting for Sparse Linear Regression
Authors: Mo Zhou, Rong Ge
ICML 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we run synthetic experiments to verify our theoretical results. We choose d from 100 to 10^6 and set n = 3 sqrt(d). |
| Researcher Affiliation | Academia | Department of Computer Science, Duke University, Durham, NC, US. |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete statement or link regarding the availability of source code for the methodology described. |
| Open Datasets | No | The paper states: "data x_i ~ N(0, I) sampled from Gaussian distribution". This indicates synthetic data generation, but no link, DOI, or formal citation for a publicly available dataset is provided. |
| Dataset Splits | No | The paper describes synthetic data generation and mentions "training loss" and "test loss" but does not specify explicit training, validation, or test dataset splits (e.g., percentages, counts, or predefined splits). |
| Hardware Specification | No | The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions "Lasso (implemented in sklearn)" but does not provide specific version numbers for sklearn or any other software dependencies, which is necessary for reproducibility. |
| Experiment Setup | Yes | We set lambda = 100d/sigma*n log(n)( sqrt(log(d)/n) + sqrt(n/d)) and run gradient descent with stepsize eta = 10^-6 until training loss reaches 10^-4. |