Deeper linear state-space models can match the expressive power of shallower wider ones, and under norm constraints they do so with exponentially smaller parameter norms.
The emergence of clusters in self-attention dynamics
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The Effect of Depth on the Expressivity of Deep Linear State-Space Models
Deeper linear state-space models can match the expressive power of shallower wider ones, and under norm constraints they do so with exponentially smaller parameter norms.