The $\mathcal{D}$-Geometric Hilbert Scheme -- Part II: Hilbert and Quot DG-Schemes

The Big Picture

Imagine you’re trying to catalog every possible physical law. Not just Newton’s equations or Maxwell’s equations in isolation, but all systems of partial differential equations (PDEs, the mathematical language describing how fields, particles, and forces evolve through space and time). Think of this catalog as a map: each point represents a different type of equation, and nearby points represent similar equations that differ by small tweaks. In mathematics, such a map is called a moduli space. Building one for PDEs has been extraordinarily difficult; the tools to do it properly barely existed until recently.

In a two-part series, Jacob Kryczka and Artan Sheshmani have been assembling exactly such a catalog, with a twist. Their version doesn’t just list equations. It tracks how one PDE system can deform into another, which systems are “stable” enough to have well-behaved solutions, and how infinitesimal changes ripple through entire families of differential equations.

Part I built the basic catalog. Part II (this paper) adds a richer layer called derived geometry. The core idea: classical geometry asks whether a mathematical object exists; derived geometry asks in how many ways it can exist, and how those ways relate to each other. That extra information (the structure of the space of possibilities, not just the possibilities themselves) is what lets mathematicians rigorously study how equations deform and interact.

Kryczka and Sheshmani prove that the space classifying a natural class of algebraic PDEs admits a full derived enhancement, representable by a concrete geometric object called a differential graded D-manifold.

Key Insight: By lifting the classical moduli space of PDEs into derived geometry, the authors create a framework where not just solutions, but the relationships between solutions, become first-class geometric objects — opening the door to a rigorous deformation theory for algebraic differential equations.

How It Works

The story begins with D-modules: algebraic objects that encode a differential equation’s structure without writing it out explicitly. Instead of working with an equation as a formula, a D-module captures it as an algebraic gadget that can be manipulated, compared, and classified. The paper studies D-ideal sheaves specifically: collections of constraints inside the algebra of infinite-order derivatives, representing the prolongations of a PDE extended to every order. Classifying these ideals up to isomorphism is the central moduli problem.

Part I established the D-Hilbert and D-Quot functors, the D-geometric analogues of Grothendieck’s classical Hilbert scheme construction. A Hilbert scheme organizes families of geometric subobjects into a single well-behaved space; the D-Quot functor does the same for families of PDE systems. The key stability condition is Spencer stability: a cohomological criterion that singles out well-behaved PDE systems, analogous to Gieseker stability for sheaves on projective varieties. Systems satisfying it form a finite-type moduli space, a classical, honest geometric object.

Part II goes deeper in three distinct moves:

• Derived enhancement. The authors construct a simplicial D-presheaf, a device for parametrizing objects up to homotopy, that tracks A∞-module structures on the characteristic modules of a PDE system. Derived geometry asks not just “does this PDE system exist?” but “in how many ways can it exist, and how do those ways relate?” The extra data controls obstructions to deformation, recording precisely where deformations can fail.

• DG-D-manifold representability. The central theorem proves that the derived D-Quot functor is representable by a differential graded D-manifold (dg-D-manifold), a geometric space with an algebraic grading compatible with D-geometry. This is the D-geometric analogue of a result by Ciocan-Fontanine and Kapranov for ordinary Quot schemes. The proof draws on Postnikov tower decompositions, Koszul duality for operadic algebras, and a careful analysis of A∞-structures on characteristic modules.

• Classical truncation. The derived object retains contact with classical geometry: stripping away the derived layers recovers the Hilbert moduli space from Part I. The derived structure is genuinely additive, enriching rather than replacing the classical picture.

The technical engine powering the main theorem involves operadic methods: classifying A∞-structures using algebras over operads, abstract combinatorial devices that organize the rules for composing operations. The proof identifies an “injectivity locus” inside the derived moduli space, then computes a colimit (a universal construction that assembles local pieces into a global object) to produce the dg-D-manifold. The authors also establish foundational results on D-geometric flatness, covering families, and Koszul duality that will likely prove useful well beyond this specific application.

Why It Matters

PDEs are everywhere in fundamental physics, from Yang-Mills equations in gauge theory to Einstein’s field equations in general relativity. But the geometric study of families of PDEs has lagged far behind the rich machinery available for algebraic varieties and coherent sheaves. Kryczka and Sheshmani’s framework changes that. By recasting PDEs as objects in derived algebraic D-geometry, they open the full toolkit of modern algebraic geometry (deformation theory, obstruction calculus, virtual fundamental classes) to the study of differential systems.

The authors point to the prospect of an enumerative algebraic D-geometry: a program that would count solutions to families of PDEs the way Gromov-Witten theory counts curves on manifolds. This has direct implications for gauge theory (counting instantons, monopoles) and complex geometry (counting special submanifolds). The derived D-Quot scheme constructed here is the moduli-theoretic foundation any such counting program would require.

Bottom Line: Kryczka and Sheshmani have built the first rigorous derived moduli space for algebraic PDEs, proving it is representable by a dg-D-manifold, a result that sets the mathematical stage for a new enumerative geometry of differential equations with applications ranging from gauge theory to AI-assisted exploration of solution spaces.