English

Linear Memory Networks

Machine Learning 2018-11-09 v1 Machine Learning

Abstract

Recurrent neural networks can learn complex transduction problems that require maintaining and actively exploiting a memory of their inputs. Such models traditionally consider memory and input-output functionalities indissolubly entangled. We introduce a novel recurrent architecture based on the conceptual separation between the functional input-output transformation and the memory mechanism, showing how they can be implemented through different neural components. By building on such conceptualization, we introduce the Linear Memory Network, a recurrent model comprising a feedforward neural network, realizing the non-linear functional transformation, and a linear autoencoder for sequences, implementing the memory component. The resulting architecture can be efficiently trained by building on closed-form solutions to linear optimization problems. Further, by exploiting equivalence results between feedforward and recurrent neural networks we devise a pretraining schema for the proposed architecture. Experiments on polyphonic music datasets show competitive results against gated recurrent networks and other state of the art models.

Keywords

Cite

@article{arxiv.1811.03356,
  title  = {Linear Memory Networks},
  author = {Davide Bacciu and Antonio Carta and Alessandro Sperduti},
  journal= {arXiv preprint arXiv:1811.03356},
  year   = {2018}
}
R2 v1 2026-06-23T05:08:50.338Z