The Everlasting Database: Statistical Validity at a Fair Price
Authors: Blake E. Woodworth, Vitaly Feldman, Saharon Rosset, Nati Srebro
NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | The paper focuses on proposing and analyzing mechanisms (VALIDATIONROUND, EVERLASTINGVALIDATION, EVERLASTINGTO) with theoretical guarantees, including lemmas (Lemma 1, 2, 3) and theorems (Theorem 1, 2, 3, 4) regarding their properties like validity, sustainability, and cost. It does not describe empirical experiments, dataset evaluations, or performance metrics from actual runs. |
| Researcher Affiliation | Academia | Blake Woodworth Toyota Technological Institute at Chicago Vitaly Feldman Saharon Rosset Tel Aviv University Nathan Srebro Toyota Technological Institute at Chicago |
| Pseudocode | Yes | Algorithm 1 VALIDATIONROUND, Algorithm 2 EVERLASTINGVALIDATION, Algorithm 3 EVERLASTINGTO |
| Open Source Code | No | The paper does not provide any statement about releasing source code for the described methods, nor does it include links to a code repository. |
| Open Datasets | No | The paper is theoretical and discusses samples from an unknown distribution D and data sets S and T conceptually, but it does not refer to or provide access information for a specific publicly available or open dataset used for empirical training. |
| Dataset Splits | No | The paper is theoretical and does not describe specific dataset splits (e.g., 80/10/10) for empirical experiments. |
| Hardware Specification | No | The paper is theoretical and does not describe any specific hardware (e.g., GPU models, CPU types) used for running experiments. |
| Software Dependencies | No | The paper is theoretical and does not mention any specific software dependencies or versions (e.g., Python 3.8, PyTorch 1.9) used for experiments. |
| Experiment Setup | No | The paper is theoretical and does not describe specific experimental setup details, such as hyperparameter values, model initialization, or training schedules. |