Independent scholarship · Open accessThe monthly digest

Indexed text attachment

topos-actions-in-machine-learning-and-ai.tex

Source for Topos Actions in Machine Learning and Artificial Intelligence

View paper · Download original file

Lines 381–480 of 1207. Source text is displayed literally.

381It follows by diagonalizing the positive semidefinite382operator and does not describe an arbitrary learned383nonlinear diffusion trajectory.384 385Suppose each incidence map is invertible. Then the edge386transport from $u$ to $v$ is387$T_e=r_{ev}^{-1}r_{eu}$, and the reverse edge transports388by $T_e^{-1}$. These maps extend to a representation of389the fundamental groupoid of the graph, regarded as a390one-dimensional complex. Indeed, the incidence category has391the vertices and edges as objects and arrows from a vertex to392its incident edges. Localizing by those arrows gives the393fundamental groupoid of the barycentric subdivision. By the394universal property of localization, precisely the diagrams with395invertible incidence maps extend. Restricting to the original396vertices gives the edge transports $T_e$. Global sections correspond to397parallel families $x_v=T_e x_u$. On each connected398component choose a base vertex $b$. Transport of $x_b$399determines the whole family, and it is independent of400the chosen paths precisely when $x_b$ is fixed by every401based loop transport. Hence the global sections on a402connected graph identify with the holonomy-invariant403subspace of $V_b$. The assertion concerns graph paths.404If two-cells are added and their boundary loops are405imposed as relations, their transports must satisfy the406additional flatness conditions.407 408A rank-one example is a triangle with two positive unit409transports and one negative unit transport. The product410around the cycle is $-1$, so the only real global section411is zero. By comparison, using three positive unit412transports gives the line of constant sections. The413underlying incidence graph is unchanged. This separates414the effect of the transport from that of connectivity.415When the incidence maps are isometries, each residual416has norm $\|x_v-T_e x_u\|$. In that case the operator is417a discrete connection Laplacian in the sense relevant418to vector diffusion maps \cite{singer-wu-vdm}. For orthogonal419changes of stalk coordinates, let $Q_0$ and $Q_1$ be the block420diagonal changes at vertices and edges, respectively. The new421incidence matrices are $r'_{ev}=Q_e r_{ev}Q_v^{-1}$, giving422$\delta'=Q_1\delta Q_0^{-1}$ and $L'=Q_0 L Q_0^{-1}$.423Thus the fixed diffusion is compatible with these changes of424frames even when the incidence maps are noninvertible.425 426The same argument constrains learnable linear parameters.427For two representations $V,W$ of the graph path428groupoid, a family of local linear maps429$A_v:V_v\to W_v$ is compatible with transport exactly430when431\[432  A_v T_e^V = T_e^W A_u,\qquad e:u\to v .433\]434Equivalently, $A$ is a parallel section of the bundle of435linear maps, with transport436$A\mapsto T_e^W A(T_e^V)^{-1}$. On a connected graph437its parameter space is therefore the space of438intertwiners of the two holonomy representations at439one vertex. For sign and trivial line holonomies440around the same cycle, a scalar parameter is forced441to vanish. For two copies of sign holonomy, any scalar442at the chosen vertex extends. Weight tying here is443dictated by representation type as well as by444the connectivity of the data. The invertibility445hypothesis matters: ordinary cellular sheaves also446permit noninvertible incidence maps and unequal stalk447dimensions, for which the incidence equation still448makes sense and there need not be transport along449groupoid arrows.450 451\subsection{From global groups to local bisections}452 453Global group actions determine the shared weights in454group-equivariant convolutions \cite{cohen-welling}.455Cohen, Geiger, and Weiler characterize equivariant456linear maps between fields on homogeneous spaces by457their convolution kernels \cite{cohen-geiger-weiler}.458On meshes, de Haan et al.\ use parallel transport to459define orientation-sensitive convolutions that respect460changes of local frame \cite{dehaan-mesh}.461 462For a finite data set carrying a group action, let463$K_{x,y}:V_y\to W_x$ be the blocks of a linear464layer and let $\rho_V(g;y)$ and $\rho_W(g;x)$ be the465feature transports. Equivariance is equivalent to466\[467 K_{gx,gy}468 =\rho_W(g;x)K_{x,y}\rho_V(g;y)^{-1}.469\]470It follows by comparing coefficients of a feature471supported at $y$. Each diagonal pair orbit has one472free block in the intertwiner space of its joint473stabilizer, since two transports from a representative474to the same pair differ by a stabilizer element.475Conversely an intertwiner at each representative476propagates to a well-defined equivariant kernel.477 478Ibort, Jimenez-Vazquez, and Perez-Pardo extend this type479of kernel analysis to groupoid representations with a480chosen pseudogroup of local bisections, measure, and