Conditional Gradient Methods via Stochastic Path-Integrated Differential Estimator
Authors: Alp Yurtsever, Suvrit Sra, Volkan Cevher
ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | Our focus in this paper is on the theoretical complexity of stochastic and finite-sum FW, with an aim to identify and present the tightest results known so far. To this end, we also propose a class of novel variance-reduced stochastic optimization algorithms, based on the recent stochastic pathintegrated differential estimator technique (SPIDER) of Fang et al. (2018). ... Developing a well-tuned implementation, including one that incorporates parallel optimization, is an important piece of future work. |
| Researcher Affiliation | Academia | 1Ecole Polytechnique F ed erale de Lausanne, Switzerland 2Massachusetts Institute of Technology, USA. |
| Pseudocode | Yes | Algorithm 1 Frank-Wolfe algorithm Algorithm 2 SPIDER Frank-Wolfe Algorithm 3 SPIDER Conditional Gradient Sliding |
| Open Source Code | No | The paper does not provide any statement or link regarding the availability of open-source code for the methodology described. |
| Open Datasets | No | The paper is theoretical and does not conduct experiments with specific datasets; therefore, no information about publicly available training datasets is provided. |
| Dataset Splits | No | The paper is theoretical and does not conduct experiments with specific datasets; therefore, no training/test/validation dataset splits are provided. |
| Hardware Specification | No | The paper is theoretical and does not report on empirical experiments; therefore, no hardware specifications are mentioned. |
| Software Dependencies | No | The paper is theoretical and does not report on empirical experiments; therefore, no specific software dependencies with version numbers are mentioned. |
| Experiment Setup | No | The paper is theoretical and does not report on empirical experiments; therefore, no experimental setup details like hyperparameters or training settings are provided. |