Indexed text attachment
differential-topology-of-parametrized-neural-networks.tex
Source for Differential Topology of Parametrized Neural Networks
View paper · Download original file
Lines 514–613 of 1074. Source text is displayed literally.
514ambient chart at a point of the circle. The equations515are $t=0$ and $tc=0$. The ordinary ideal is $(t)$,516so the ordinary fiber has a smooth one-dimensional517function ring on this chart.518 519For the derived fiber, use the smooth Koszul model520\cite[Chs.~21--22]{higher-hirsch}.521Its cochain algebra is $A\otimes\bigwedge(\eta_1,\eta_2)$522with $|\eta_i|=-1$, $d\eta_1=t$, and $d\eta_2=tc$.523The invertible change $\eta_2'=\eta_2-c\eta_1$ gives524$d\eta_2'=0$. Multiplication by $t$ on smooth germs is525injective because $t\ne0$ on a dense open subset of the526chart. The first generator thus resolves $A/(t)$ and527the second gives cohomology528\[529 H^0=A/(t),\qquad530 H^{-1}=(A/(t))\eta_2'.531\]532The remaining negative cohomology vanishes. The generator533$\eta_2'$ transforms by sign. The tangent calculation gives534the same excess on the dual side:535\[536 T_x^{\rm der}537 =\bigl[\,538 \mathbb R_-\oplus\mathbb R_+539 \xrightarrow{\ (a,b)\mapsto(a,ca)\ }540 \mathbb R_-^2541 \,\bigr].542\]543The terms have cohomological degrees zero and one.544The kernel is the trivial tangent line of the fixed circle.545The quotient of the target by the image is the sign line,546explicitly identified by $(u_1,u_2)\mapsto u_2-cu_1$.547The virtual dimension is zero. The ordinary dimension548is one, and the derived equation record specifies the excess549direction and its stabilizer action.550 551The squared readout loss provides another comparison.552With the usual invariant Euclidean metric on the outputs,553its local expression is554\[555 \ell(t,\phi)=\tfrac12\|f(t,\phi)\|^2556 =\tfrac12 t^2(1+\cos^2\phi).557\]558At $t=0$ the Hessian has normal eigenvalue559$1+\cos^2\phi$ and tangential eigenvalue zero in the560chosen coordinate basis. Thus each fixed circle is a561smooth critical manifold with positive-definite Hessian562on its normal line. This loss detects movement away from563the circle. The excess target line survives even though564the loss controls every normal displacement. At a zero,565the general identity $D^2\ell=(Df)^*Df$ identifies the566null directions of the Hessian with $\ker Df$, whereas567the derived tangent complex encodes the cokernel in a568separate degree.569 570\subsection{A moving bias and the symmetric fiber}571 572Take a bias $b\in P=V$ transforming by the same sign action,573and set $F(b,x)=f(x)-b$. The total zero equation is regular574because $D_bF=-I_V$. Its ordinary total incidence space575$Z=\{(b,x):b=f(x)\}$ is the smooth graph of $f$.576Projection $Z\to P$ identifies with $f:X\to V$ and is577singular along its fiber over zero. The parameter $b=0$578has stabilizer $G$, and the readout on this slice is579exactly the original rank-deficient equivariant map.580In contrast, if $0<|b_1|<1$ and $|b_2/b_1|<1$,581the fiber has four points. The equations prescribe582$\sin\theta=b_1$ and $\cos\phi=b_2/b_1$, with two choices583for each angle. The derivative determinant is584$-\cos\theta\sin\theta\sin\phi$, which is nonzero at585all four points. At such a bias the parameter stabilizer586is trivial.587 588The quotient also retains the type of the parameter.589For this finite group, the smooth action stacks give590a map $[Z/G]\to[P/G]$ as in591\cite[Ch.~17]{higher-hirsch}. The zero-parameter orbit592in the base is $BG=[\{0\}/G]$. In the classical smooth593setting the pullback over this orbit is $[C/G]$,594where $C$ is the pair of fixed circles. It has the595product description $C\times BG$. In derived smooth596stacks the homotopy pullback over that orbit is597$[Z_0^{\rm der}/G]$, with $Z_0^{\rm der}$ the derived598zero fiber computed above. Indeed, pulling back along599the atlas of $BG$ trivializes the torsor and gives600the derived incidence square on the manifolds. Gluing601this calculation under the torsor's changes of frame602recovers its $G$-action. Thus the total evaluation can603be made regular by the bias, while its symmetric604fiber keeps both the stabilizer and the excess605equation of the fixed-parameter model.606\par607\section{Decision boundaries with parameter monodromy}608\label{sec:boundary-monodromy}609 610The regular fibers of a score can carry a permutation611representation of the parameter space's fundamental group.612An invariant computed separately at each parameter can miss613this information. We apply the signed-fiber, relative mapping,