A Generalized Framework for Population Based Training

Ang Li

[0]

[0]

[0]

[0]

[0]

[0]

[0]

[0]

[0]

KDD, pp. 1791.0-1799.0, 2019.

Cited by: 12|Views136

Keywords:

black-box optimization evolutionary algorithms neural networks population based training speech synthesisMore(1+)

Weibo:

We show that the proposed Population Based Training framework is feasible for large scale deep neural network training

Abstract:

Population Based Training (PBT) is a recent approach that jointly optimizes neural network weights and hyperparameters which periodically copies weights of the best performers and mutates hyperparameters during training. Previous PBT implementations have been synchronized glass-box systems. We propose a general, black-box PBT framework th...More

Code:

Data:

Full Text

View via Publisher