Diagonal State Spaces are as Effective as Structured State Spaces

Ankit Gupta, Albert Gu, Jonathan Berant

Keywords: state spaces, long range models, efficient, Transformer, speech recognition, language modeling, time series model

TL;DR: We propose a seq2seq model that uses diagonal state spaces (DSS) for contextualization & delivers state-of-the-art performance on benchmarks requiring long-range reasoning over text, images & audio.

Abstract: Modeling long range dependencies in sequential data is a fundamental step towards attaining human-level performance in many modalities such as text, vision, audio and video. While attention-based models are a popular and effective choice in modeling short-range interactions, their performance on tasks requiring long range reasoning has been largely inadequate. In an exciting result, Gu et al. (ICLR 2022) proposed the $\textit{Structured State Space}$ (S4) architecture delivering large gains over state-of-the-art models on several long-range tasks across various modalities. The core proposition of S4 is the parameterization of state matrices via a diagonal plus low rank structure, allowing efficient computation. In this work, we show that one can match the performance of S4 even without the low rank correction and thus assuming the state matrices to be diagonal. Our $\textit{Diagonal State Space}$ (DSS) model matches the performance of S4 on Long Range Arena tasks, speech classification on Speech Commands dataset, while being conceptually simpler and straightforward to implement.

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:2203.14343/code)