Corpus ID: 29233887

Underestimate Sequences via Quadratic Averaging

@article{Ma2017UnderestimateSV,
  title={Underestimate Sequences via Quadratic Averaging},
  author={Chenxin Ma and N. V. C. Gudapati and Majid Jahani and R. Tappenden and Martin Tak{\'a}c},
  journal={ArXiv},
  year={2017},
  volume={abs/1710.03695}
}
In this work we introduce the concept of an Underestimate Sequence (UES), which is a natural extension of Nesterov's estimate sequence. Our definition of a UES utilizes three sequences, one of which is a lower bound (or under-estimator) of the objective function. The question of how to construct an appropriate sequence of lower bounds is also addressed, and we present lower bounds for strongly convex smooth functions and for strongly convex composite functions, which adhere to the UES framework… Expand

References

SHOWING 1-10 OF 28 REFERENCES
Estimate sequence methods: extensions and approximations
Gradient methods for minimizing composite objective function
Gradient methods for minimizing composite functions
  • Y. Nesterov
  • Mathematics, Computer Science
  • Math. Program.
  • 2013
An Optimal First Order Method Based on Optimal Quadratic Averaging
Minimizing finite sums with the stochastic average gradient
On the complexity analysis of randomized block-coordinate descent methods
A Universal Catalyst for First-Order Optimization
Iteration complexity of randomized block-coordinate descent methods for minimizing a composite function
Accelerated, Parallel, and Proximal Coordinate Descent
...
1
2
3
...