Search papers, labs, and topics across Lattice.
This paper introduces a geometric parametrization of convolutional layer filters by representing them as fixed-dimensional subspaces rather than ordered families of vectors. The authors establish an injective linear map that connects filter subspaces to operator subspaces, leading to a projective parametrization that is shown to be a closed embedding. Key findings include the smoothness of the resulting projective neural variety and implications for filter redundancy and low-rank convolution, highlighting the theoretical robustness of the approach.
The proposed geometric framework reveals that convolutional filter subspaces can be parametrized as smooth projective varieties, offering new insights into filter redundancy and low-rank structures.
We propose a geometric parametrization of the filters in a single convolutional layer: the parameter is no longer an ordered family of filter vectors, but a fixed-dimensional subspace of the filter space. For one-dimensional finite-stride convolution, the filter-to-convolution-operator correspondence gives an injective linear map $\mathcal{C}:\mathcal{K}\to H$. This map sends filter subspaces in $\mathrm{Gr}(q,\mathcal{K})$ to operator subspaces in $\mathrm{Gr}(q,H)$; composing it with the Pl\"ucker embedding yields a projective parametrization $\Phi:\mathrm{Gr}(q,\mathcal{K})\to\mathbb{P}(\bigwedge^q H)$. Using $T_U\mathrm{Gr}(q,\mathcal{K})\cong\mathrm{Hom}(U,\mathcal{K}/U)$, we compute the differential of the induced Grassmannian map and show that the differential of $\Phi$ is injective at every point. We then use the vanishing equations for Pl\"ucker coordinates and standard affine coordinates on a Grassmannian to prove that $\mathrm{Gr}(q,\mathcal{C}(\mathcal{K}))\hookrightarrow\mathrm{Gr}(q,H)$ is a closed embedding, and hence that $\Phi$ is a closed embedding. Consequently, the parameter space is isomorphic to its projective image, the parametrization is finite and birational onto its image, every fiber is a singleton, and the resulting projective neural variety is smooth. For $k=4$ and $q=2$, we also use Singular to recover the image ideal and check its dimension, degree, chart rank, and smoothness. This computation illustrates, rather than replaces, the general proof. Finally, we discuss possible connections with filter redundancy and low-rank convolution, while distinguishing the proved geometric results from application proposals requiring numerical validation.
