Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Ridge Boosting is Both Robust and Efficient
Authors: David Bruns-Smith, Zhongming Xie, Avi Feller
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We assess this approach through simulations and an application estimating the age profile of retirement income. (...) 4 Experiments 4.1 Simulation study 4.2 Empirical application to retirement income |
| Researcher Affiliation | Academia | David Bruns-Smith (Stanford University) Zhongming Xie (UC Berkeley) Avi Feller (UC Berkeley) |
| Pseudocode | No | The paper describes the methodology using mathematical equations and prose but does not include a distinct, labeled pseudocode or algorithm block. |
| Open Source Code | Yes | 5. Open access to data and code Question: Does the paper provide open access to the data and code, with sufficient instructions to faithfully reproduce the main experimental results, as described in supplemental material? Answer: [Yes] Justification: The code is provided in the Supplemental Material. |
| Open Datasets | Yes | We estimate the age profile of retirement using data from 2018 American Community Survey as processed by the Folk Tables package [Ding et al., 2021]. |
| Dataset Splits | Yes | Monte Carlo Simulation: For each simulation, we draw a training sample of X and Y from the source distribution, and fit the kernel ridge and boosted kernel ridge model. We then draw one sample of X from each of the three test distributions, and estimate the average derivative with respect to X1 on that test distribution by symmetric differencing, along with the usual 95% asymptotic normal confidence interval. The whole process is repeated for sample sizes ranging from 50 to 500 and with 1,000 Monte Carlo replications. |
| Hardware Specification | Yes | The full simulation study is run on a four-core laptop. |
| Software Dependencies | No | The paper does not explicitly mention specific software dependencies with version numbers. |
| Experiment Setup | Yes | The ridge boosting estimator with λ = n^-1/2 is asymptotically normal and its variance achieves the asymptotic variance lower bound V θ. |