An overview of gradient descent optimization algorithms

Explore this paper's citation graph

Summary

A technique is proposed, when a test sample is divided into N parts, on each of which the values of the metrics are calculated, which reduces the amount of computations needed, however, an experimental analysis of the binary cross-entropy metric for CTR prediction models showed that it is more rough than the first one.

Type
preprint
Published
2016-09-15
Cited by
7,037
References
31
Access
Open access

Keywords

Gradient descent, Descent (aeronautics), Algorithm, Computer science, Optimization algorithm

References

Cited by

Related papers