Spectral state space models

[Free GPT-4]
[PDF] arxiv.org

State space model for new-generation network alternative to transformers: A survey

X Wang, S Wang, Y Ding, Y Li, W Wu, Y Rong… - arxiv preprint arxiv …, 2024 - arxiv.org
In the post-deep learning era, the Transformer architecture has demonstrated its powerful
performance across pre-trained big models and various downstream tasks. However, the …
保存 引用 被引用数: 38 関連記事 全 2 バージョン HTMLバージョン
[Free GPT-4]
[PDF] arxiv.org

Tuning frequency bias of state space models

A Yu, D Lyu, SH Lim, MW Mahoney… - arxiv preprint arxiv …, 2024 - arxiv.org
State space models (SSMs) leverage linear, time-invariant (LTI) systems to effectively learn
sequences with long-range dependencies. By analyzing the transfer functions of LTI …
保存 引用 被引用数: 2 関連記事 全 3 バージョン HTMLバージョン
[Free GPT-4]
[PDF] arxiv.org

Computation-Efficient Era: A Comprehensive Survey of State Space Models in Medical Image Analysis

M Heidari, SG Kolahi, S Karimijafarbigloo… - arxiv preprint arxiv …, 2024 - arxiv.org
Sequence modeling plays a vital role across various domains, with recurrent neural
networks being historically the predominant method of performing these tasks. However, the …
保存 引用 被引用数: 7 関連記事 全 2 バージョン HTMLバージョン
[Free GPT-4]
[PDF] arxiv.org

There is HOPE to Avoid HiPPOs for Long-memory State Space Models

A Yu, MW Mahoney, NB Erichson - arxiv preprint arxiv:2405.13975, 2024 - arxiv.org
State-space models (SSMs) that utilize linear, time-invariant (LTI) systems are known for
their effectiveness in learning long sequences. However, these models typically face several …
保存 引用 被引用数: 5 関連記事 全 2 バージョン HTMLバージョン
[Free GPT-4]
[PDF] arxiv.org

Flash stu: Fast spectral transform units

YI Liu, W Nguyen, Y Devre, E Dogariu… - arxiv preprint arxiv …, 2024 - arxiv.org
This paper describes an efficient, open source PyTorch implementation of the Spectral
Transform Unit. We investigate sequence prediction tasks over several modalities including …
保存 引用 被引用数: 1 関連記事 全 2 バージョン HTMLバージョン
[Free GPT-4]
[PDF] arxiv.org

A Deep State Space Model for Rainfall-Runoff Simulations

Y Wang, L Zhang, A Yu, NB Erichson… - arxiv preprint arxiv …, 2025 - arxiv.org
The classical way of studying the rainfall-runoff processes in the water cycle relies on
conceptual or physically-based hydrologic models. Deep learning (DL) has recently …
保存 引用 関連記事 HTMLバージョン
[Free GPT-4]
[PDF] arxiv.org

Provable Length Generalization in Sequence Prediction via Spectral Filtering

A Marsden, E Dogariu, N Agarwal, X Chen… - arxiv preprint arxiv …, 2024 - arxiv.org
We consider the problem of length generalization in sequence prediction. We define a new
metric of performance in this setting--the Asymmetric-Regret--which measures regret against …
保存 引用 関連記事 HTMLバージョン
[Free GPT-4]
[PDF] arxiv.org

FutureFill: Fast Generation from Convolutional Sequence Models

N Agarwal, X Chen, E Dogariu, V Feinberg… - arxiv preprint arxiv …, 2024 - arxiv.org
We address the challenge of efficient auto-regressive generation in sequence prediction
models by introducing FutureFill-a method for fast generation that applies to any sequence …
保存 引用 関連記事 HTMLバージョン