## Abstract

Many canonical machine learning problems boil down to a convex optimization problem with a finite sum structure. However, whereas much progress has been made in developing faster algorithms for this setting, the inherent limitations of these problems are not satisfactorily addressed by existing lower bounds. Indeed, current bounds focus on first-order optimization algorithms, and only apply in the often unrealistic regime where the number of iterations is less than O(d/n) (where d is the dimension and n is the number of samples). In this work, we extend the framework of Arjevani et al. [3, 5] to provide new lower bounds, which are dimension-free, and go beyond the assumptions of current bounds, thereby covering standard finite sum optimization methods, e.g., SAG, SAGA, SVRG, SDCA without duality, as well as stochastic coordinate-descent methods, such as SDCA and accelerated proximal SDCA.

Original language | English |
---|---|

Pages (from-to) | 3548-3555 |

Number of pages | 8 |

Journal | Advances in Neural Information Processing Systems |

State | Published - 2016 |

Externally published | Yes |

Event | 30th Annual Conference on Neural Information Processing Systems, NIPS 2016 - Barcelona, Spain Duration: 5 Dec 2016 → 10 Dec 2016 |

### Bibliographical note

Publisher Copyright:© 2016 NIPS Foundation - All Rights Reserved.