arxivst stuff from arxiv that you should probably bookmark

Learned Optimizers that Scale and Generalize

Abstract · Mar 14, 2017 23:05 ·

cs-lg cs-ne stat-ml

Arxiv Abstract

  • Olga Wichrowska
  • Niru Maheswaranathan
  • Matthew W. Hoffman
  • Sergio Gomez Colmenarejo
  • Misha Denil
  • Nando de Freitas
  • Jascha Sohl-Dickstein

Learning to learn has emerged as an important direction for achieving artificial intelligence. Two of the primary barriers to its adoption are an inability to scale to larger problems and a limited ability to generalize to new tasks. We introduce a learned gradient descent optimizer that generalizes well to new tasks, and which has significantly reduced memory and computation overhead. We achieve this by introducing a novel hierarchical RNN architecture, with minimal per-parameter overhead, augmented with additional architectural features that mirror the known structure of optimization tasks. We also develop a meta-training ensemble of small, diverse, optimization tasks capturing common properties of loss landscapes. The optimizer learns to out-perform RMSProp/ADAM on problems in this corpus. More importantly, it performs comparably or better when applied to small convolutional neural networks, despite seeing no neural networks in its meta-training set. Finally, it generalizes to train Inception V3 and ResNet V2 architectures on the ImageNet dataset, optimization problems that are of a vastly different scale than those it was trained on.

Read the paper (pdf) »