Efficiently Modeling Long Sequences with Structured State Spaces

ICLR 2022  ·  Albert Gu, Karan Goel, Christopher Ré ·

A central goal of sequence modeling is designing a single principled model that can address sequence data across a range of modalities and tasks, particularly on long-range dependencies. Although conventional models including RNNs, CNNs, and Transformers have specialized variants for capturing long dependencies, they still struggle to scale to very long sequences of $10000$ or more steps. A promising recent approach proposed modeling sequences by simulating the fundamental state space model (SSM) \( x'(t) = Ax(t) + Bu(t), y(t) = Cx(t) + Du(t) \), and showed that for appropriate choices of the state matrix \( A \), this system could handle long-range dependencies mathematically and empirically. However, this method has prohibitive computation and memory requirements, rendering it infeasible as a general sequence modeling solution. We propose the Structured State Space sequence model (S4) based on a new parameterization for the SSM, and show that it can be computed much more efficiently than prior approaches while preserving their theoretical strengths. Our technique involves conditioning \( A \) with a low-rank correction, allowing it to be diagonalized stably and reducing the SSM to the well-studied computation of a Cauchy kernel. S4 achieves strong empirical results across a diverse range of established benchmarks, including (i) 91\% accuracy on sequential CIFAR-10 with no data augmentation or auxiliary losses, on par with a larger 2-D ResNet, (ii) substantially closing the gap to Transformers on image and language modeling tasks, while performing generation $60\times$ faster (iii) SoTA on every task from the Long Range Arena benchmark, including solving the challenging Path-X task of length 16k that all prior work fails on, while being as efficient as all competitors.

PDF Abstract ICLR 2022 PDF ICLR 2022 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Long-range modeling LRA S4 ListOps 58.35 # 1
Text 76.02 # 5
Retrieval 87.09 # 2
Image 87.26 # 1
Pathfinder 86.05 # 2
Avg 80.48 # 1
Pathfinder-X 88.1 # 1
Sequential Image Classification Sequential CIFAR-10 S4 Unpermuted Accuracy 91.13% # 1
Sequential Image Classification Sequential MNIST S4 Unpermuted Accuracy 99.63% # 1
Permuted Accuracy 98.70% # 3
Language Modelling WikiText-103 S4 Test perplexity 21.28 # 27
Number of params 249M # 13

Methods