StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization.

AllImages Books Videos Maps News Shopping

StableSSM: Alleviating the Curse of Memory in State-space Models ...

... curse of memory" as a result of the recurrent weights converging to a ... stability. We validate our findings using synthetic datasets and language models.

Collections - Hugging Face

huggingface.co › collections

Collections · StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization · Vision Mamba: Efficient Visual Representation ...

Qianxiao Li | Papers With Code

paperswithcode.com › author › qianxiao-li

StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization · 1 code implementation • 24 Nov 2023 • Shida Wang ...

Wang Shida's Personal Website

radarfudan.github.io

GitHub repo: Curse-of-memory · 2. StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization (ICML 2024) · 3. State- ...

State-space Models with Layer-wise Nonlinearity are ... - arxiv-sanity

www.arxiv-sanity-lite.com › ...

Our analysis identifies this "curse of memory" as a result of the recurrent weights converging to a stability boundary, suggesting that a reparameterization ...

Qianxiao Li - CatalyzeX

www.catalyzex.com › author

Our analysis identifies this "curse of memory" as a result of the recurrent weights converging to a stability boundary, suggesting that a reparameterization ...

State-space Models with Layer-wise Nonlinearity are Universal ...

www.semanticscholar.org › paper › State...

Theoretical Foundations of Deep Selective State-Space Models · StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization.

LongSSM: On the Length Extension of State-space Models in Language ...

www.researchgate.net › publication › 38...

Jun 7, 2024 · StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization. Nov 2023. Shida Wang; Qianxiao Li. Shida ...

Recurrent neural networks: vanishing and exploding gradients are not the ...

www.aimodels.fyi › papers › arxiv › rec...

Jun 2, 2024 · StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization · Recurrent Natural Policy Gradient for POMDPs.

On the Curse of Memory in Recurrent Neural Networks: Approximation ...

www.semanticscholar.org › paper › On-t...

StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization · Shida WangQianxiao Li. Computer Science. ArXiv. 2023. TLDR.