Improving the Neural GPU Architecture for Algorithm Learning

Anonymous

May 29, 2018 (modified: Jul 04, 2018) ICML 2018 Workshop NAMPI Blind Submission readers: everyone
  • Abstract: Algorithm learning is a core problem in artificial intelligence with significant implications on automation level that can be achieved by machines. Recently deep learning methods are emerging for synthesizing an algorithm from its input-output examples, the most successful being the Neural GPU, capable of learning multiplication. We present several improvements to the Neural GPU that substantially reduces training time and improves generalization. We introduce a new technique - hard nonlinearities with saturation costs - that has general applicability. We also introduce a technique of diagonal gates that can be applied to active-memory models. The proposed architecture is the first capable of learning decimal multiplication end-to-end.
  • Keywords: Algorithm learning, Neural GPU, convolution
  • TL;DR: We present several improvements to the Neural GPU that substantially reduces training time and improves generalization
0 Replies

Loading