The Statistical Complexity of Early-Stopped Mirror Descent
Authors: Tomas Vaskevicius, Varun Kanade, Patrick Rebeschini
NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Figure 1: Consider a distribution P such that X N(0, Id) and Y |X = x α , x +N(0, 52) for some parameter α Rd. Fix n = 200, d = 100 and let α be a 10-sparse vector with nonzero entries equal to 1. ... In the plot above, the solid lines denote means over 100 runs whereas the shaded regions correspond to the 10th and the 90th percentiles. Figure 2: Consider the setting of Figure 1 and let εt = Rn(αt) Rn(α ) + gαt gα 2 n. The above plots illustrate the following two points. ... In the plot above, the solid lines denote means over 100 runs, the dots denote the minimum of each solid line, whereas the shaded regions correspond to the 10th and the 90th percentiles. |
| Researcher Affiliation | Academia | Tomas Vaškevičius1, Varun Kanade2, Patrick Rebeschini1 1 Department of Statistics, 2 Department of Computer Science University of Oxford {tomas.vaskevicius, patrick.rebeschini}@stats.ox.ac.uk varunk@cs.ox.ac.uk |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide concrete access to source code for the methodology described. It mentions a 'full version of this paper' in reference [39], which is an arXiv preprint, not a code repository. |
| Open Datasets | No | The paper describes synthetic data generation parameters (e.g., 'Fix n = 200, d = 100', 'X N(0, Id)', 'Y |X = x α , x +N(0, 52)') but does not provide concrete access information (link, DOI, repository, or formal citation) for a publicly available or open dataset. |
| Dataset Splits | No | The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning. |
| Hardware Specification | No | The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments. |
| Software Dependencies | No | The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment. |
| Experiment Setup | No | The paper does not contain specific experimental setup details (concrete hyperparameter values, training configurations, or system-level settings) in the main text for the empirical demonstrations. |