LGJun 13, 2015
On the Equivalence of CoCoA+ and DisDCA
arXiv:1506.04217v2
Originality Synthesis-oriented
AI Analysis
This work clarifies the relationship between two existing distributed optimization algorithms for machine learning, but it is incremental as it does not introduce new methods or solve a broader problem.
The paper demonstrates that the CoCoA+ algorithm, under its recommended experimental settings, is equivalent to the practical variant of DisDCA, showing no new performance improvements or results.
In this document, we show that the algorithm CoCoA+ (Ma et al., ICML, 2015) under the setting used in their experiments, which is also the best setting suggested by the authors that proposed this algorithm, is equivalent to the practical variant of DisDCA (Yang, NIPS, 2013).