Multi-class classification: mirror descent approach
arXiv:1607.00076v2
Originality Incremental advance
AI Analysis
This work addresses multi-class classification for machine learning practitioners, presenting an incremental improvement with theoretical analysis.
The paper tackles multi-class classification by applying a stochastic mirror descent algorithm, deriving risk bounds and demonstrating efficient error rates for specific set geometries.
We consider the problem of multi-class classification and a stochastic opti- mization approach to it. We derive risk bounds for stochastic mirror descent algorithm and provide examples of set geometries that make the use of the algorithm efficient in terms of error in k.