Long Expressive Memory for Sequence Modeling

Abstract

We propose a novel method called Long Expressive Memory (LEM) for learning long-term sequential dependencies. LEM is gradient-based, it can efficiently process sequential tasks with very long-term dependencies, and it is sufficiently expressive to be able to learn complicated input-output maps. To derive LEM, we consider a system of multiscale ordinary differential equations, as well as a suitable time-discretization of this system. For LEM, we derive rigorous bounds to show the mitigation of the exploding and vanishing gradients problem, a well-known challenge for gradient-based recurrent sequential learning methods. We also prove that LEM can approximate a large class of dynamical systems to high accuracy. Our empirical results, ranging from image and time-series classification through dynamical systems prediction to speech recognition and language modeling, demonstrate that LEM outperforms state-of-the-art recurrent neural networks, gated recurrent units, and long short-term memory models.

ICLR 2022

References in corpus (8)

• On the difficulty of training Recurrent Neural Networks
• Latent ODEs for Irregularly-Sampled Time Series
• Full-Capacity Unitary Recurrent Neural Networks
• AntisymmetricRNN: A Dynamical System View on Recurrent Neural Networks
• Lipschitz Recurrent Neural Networks
• Continuous-in-Depth Neural Networks
• Monash University, UEA, UCR Time Series Extrinsic Regression Archive
• Noisy Recurrent Neural Networks