A tower of complete moduli spaces of Calabi-Yau $n$-folds

This paper constructs a sequence of complete moduli spaces isomorphic to weighted projective spaces that parameterize specific $n$-dimensional Calabi-Yau varieties associated with the Sylvester sequence, thereby generalizing the moduli of elliptic curves and Brieskorn's family of K3 surfaces while extending the theory of elliptic surfaces to higher dimensions.

The Gist

Imagine you are an architect trying to design a perfect, self-contained universe. In mathematics, these "universes" are called Calabi-Yau varieties. They are complex, multi-dimensional shapes that are crucial for string theory (the idea that the universe is made of tiny vibrating strings) and for understanding the deep symmetries of geometry.

For a long time, mathematicians have struggled to organize these shapes into a neat catalog, or "moduli space." Usually, if you try to list all possible Calabi-Yau shapes, the list gets messy, incomplete, or has holes where the shapes break down.

Valery Alexeev's paper is like building a perfect, infinite tower of catalogs that solves this problem. Here is a simple breakdown of what he did, using everyday analogies.

1. The "Sylvester Sequence": The Secret Recipe

The paper relies on a specific list of numbers called the Sylvester sequence: 2, 3, 7, 43, 1807, and so on.

• The Analogy: Imagine you are baking a cake. Most recipes use standard cups and spoons. But this recipe uses a very specific, weird set of measuring cups that grow incredibly fast.
• The Magic: If you use these specific numbers to define the "weights" of the ingredients (the variables in the math equation), something magical happens. The resulting shapes (Calabi-Yau varieties) become incredibly well-behaved. They don't break; they fit together perfectly.

2. The Tower of Spaces ($E_0, E_1, E_2, \dots$)

Alexeev constructs a sequence of spaces, $E_0 \subset E_1 \subset E_2 \dots$.

• The Analogy: Think of these as floors in a skyscraper.
  • Floor 0 ($E_0$): A tiny room.
  • Floor 1 ($E_1$): A slightly bigger room that contains Floor 0. This floor is actually the famous "moduli space of elliptic curves" (the shapes of donuts).
  • Floor 2 ($E_2$): A massive hall containing Floor 1. This floor catalogs "K3 surfaces" (complex 2D shapes).
  • Floor $n$: A gigantic, multi-dimensional space that catalogs Calabi-Yau shapes of dimension $n$.
• The Catch: As you go up the tower, the rooms get astronomically huge. By the time you reach the 5th floor, the number of dimensions is larger than the number of atoms in the universe! But despite their size, they are all "complete," meaning they have no holes or missing pieces.

3. The "Short Weierstrass Equation": The Universal Blueprint

To build these shapes, Alexeev uses a specific type of equation, which he calls a "short Weierstrass equation."

• The Analogy: Imagine a Lego set. Usually, if you want to build different castles, you need a million different bricks. But here, Alexeev found a "Master Brick" formula.
• By tweaking a few knobs (the coefficients $t_I$ in the equation), you can generate any shape in that specific family. It's like having a single 3D printer blueprint where you just change a few settings to print a chair, a table, or a house, and they all fit perfectly into the same catalog.

4. The "Boundary" and the "Smoothing"

One of the hardest parts of geometry is dealing with shapes that have sharp corners or cracks (singularities).

• The Problem: Usually, when a shape gets too weird, it falls out of the catalog.
• The Solution: Alexeev shows that even the "broken" shapes at the edge of his tower (the boundary) are actually just deformed versions of the perfect shapes inside.
• The Analogy: Think of a clay sculpture. If you squish it too hard, it cracks. But in this tower, even the cracked clay is considered a valid, stable part of the collection. The "boundary" is just the limit where the clay is about to break, but it's still part of the same family.

5. Why This Matters (The "Local Torelli" Theorem)

The paper proves that if you know the "periods" (a mathematical fingerprint) of these shapes, you can uniquely identify the shape itself.

• The Analogy: Imagine you have a bag of unique snowflakes. Usually, if two snowflakes look similar, they might still be different. But Alexeev proves that for these specific shapes, if their fingerprints match, they are the exact same snowflake.
• This is a huge deal because it means the catalog is not just a list; it's a perfect map. If you know the coordinates in the catalog, you know the shape exactly.

Summary

Valery Alexeev has built a mathematical skyscraper where every floor is a complete, perfect catalog of complex geometric shapes.

1. He used a weird number sequence (2, 3, 7, 43...) to make the shapes behave.
2. He created a universal blueprint (the equation) that generates all these shapes.
3. He proved that even the broken or "cracked" shapes at the edges of the catalog are stable and fit in perfectly.
4. He showed that these catalogs are rigid and precise: if you know the math fingerprint, you know the shape.

It's like taking a chaotic pile of puzzle pieces and realizing they all fit into a single, infinite, perfectly organized box, where every piece, no matter how weird, has its exact place.

Technical Summary

1. Problem Statement

The paper addresses the construction and classification of complete moduli spaces for Calabi-Yau varieties of dimension $n$. While moduli spaces for curves ($n=1$) and K3 surfaces ($n=2$) are well-understood (e.g., the moduli of elliptic curves $M_{1,1}$ and the Baily-Borel compactification for K3 surfaces), constructing explicit, complete moduli spaces for higher-dimensional Calabi-Yau varieties ($n \geq 3$) remains a significant challenge.

The author aims to:

1. Construct a sequence of complete moduli spaces $E_0 \subset E_1 \subset E_2 \subset \dots \subset E_n \subset \dots$.
2. Show that these spaces are isomorphic to weighted projective spaces.
3. Parameterize specific Calabi-Yau varieties associated with the Sylvester sequence ($2, 3, 7, 43, \dots$).
4. Establish the singularity types (canonical, log canonical) of the fibers and prove a local Torelli theorem.

2. Methodology and Mathematical Framework

The Sylvester Sequence and Weights

The construction relies on the Sylvester sequence $\{s_k\}$, defined recursively by $s_0 = 2$ and $s_{k+1} = 1 + \prod_{i=0}^k s_i$.

• The sequence begins: $2, 3, 7, 43, 1807, \dots$.
• Let $d_n = \prod_{k=0}^n s_k$ and $a_{n,k} = d_n / s_k$.
• These numbers satisfy the Egyptian fraction identity: $\sum_{k=0}^n \frac{1}{s_k} + \frac{1}{d_n} = 1$.

The Family of Hypersurfaces

The author considers Brieskorn-Pham hypersurface singularities defined by $f_0 = \sum_{k=0}^n x_k^{s_k} = 0$ in $\mathbb{C}^{n+1}$.

• Weighted Projective Space: The varieties are embedded in a weighted projective space $P = \mathbb{P}(a_0, \dots, a_n, 1)$.
• Smoothing Component: The author focuses on the "smoothing component" of the deformation space. By adding a variable $x_{n+1}$ of degree 1, the affine equation is homogenized to a degree $d$ hypersurface:
  $$F_t = \sum_{k=0}^n x_k^{s_k} + \sum_{I \in \mathcal{I}_n} t_I x^I \cdot x_{n+1}^{\delta_I} = 0$$
  where $\delta_I = d - \sum a_k i_k > 0$.
• Moduli Space Construction: The coefficients $t_I$ are rescaled by a $\mathbb{G}_m$-action. The resulting quotient stack is a weighted projective stack $E_n = \mathcal{P}(\delta_I)$, and its coarse moduli space is the weighted projective space $E_n = \mathbb{P}(\delta_I)$.

Singularity Analysis

The paper employs techniques from the Minimal Model Program (MMP):

• Newton Polyhedra: Used to analyze non-degenerate singularities.
• Inversion of Adjunction: Used to relate the singularities of the total space to the fibers.
• Fibration Structure: The author proves that the affine part of a Calabi-Yau variety in $E_n$ fibers over $\mathbb{A}^1$ with fibers isomorphic to varieties in $E_{n-1}$. This allows for inductive proofs regarding singularity types.

3. Key Contributions and Results

A. Construction of the Tower $E_n$

The paper constructs a sequence of complete moduli spaces $E_n$ where:

• $E_0 = \mathbb{P}(2)$ (a point).
• $E_1 = \mathbb{P}(4, 6) \cong \overline{M}_{1,1}$ (the moduli of elliptic curves).
• $E_2 = \mathbb{P}(4, 10, \dots, 42)$, which coincides with the Baily-Borel compactification of the moduli space of $U \oplus E_8$-polarized K3 surfaces.
• For $n \geq 3$, $E_n$ provides a new, explicit compactification for Calabi-Yau $n$-folds.
• Dimensions: The dimensions grow doubly-exponentially (e.g., $\dim E_3 = 251$, $\dim E_4 \approx 1.5 \times 10^5$).

B. Singularity Classification (Theorems 1 & 2)

The paper rigorously classifies the singularities of the fibers $(X, D)$, where $D$ is the "infinite divisor" ($x_{n+1}=0$):

1. General Fibers: For a general point in $E_n$, the variety $X$ has canonical singularities in the finite part ($X \setminus D$) and specific quotient singularities along $D$. The pair $(X, D)$ is log canonical.
2. Boundary Fibers: Points on the boundary $\partial E_n = \iota_{n-1}(E_{n-1})$ correspond to varieties with strictly log canonical singularities (if $n \neq 1$).
3. Non-Log Canonical Base: The original Brieskorn singularity $f_0$ (corresponding to the "cuspidal" fiber) is not log canonical, but its deformations in the smoothing component are.

C. Moduli Interpretations (Theorems 3 & 4)

The author provides two distinct interpretations of $E_n$:

1. Fermat-Nondegenerate Hypersurfaces: $E_n$ is the moduli stack of Fermat-nondegenerate hypersurfaces of degree $d$ in $P$ modulo the automorphism group of $(P, D)$.
2. KSBA-Stable Pairs: $E_n$ is a connected component of the moduli stack of KSBA-stable pairs $(X, D)$ (Kollár-Shepherd-Barron-Alexeev). This is significant because it proves these Calabi-Yau varieties form a proper, compact moduli space in the sense of stable pairs, despite being non-normal in the boundary (though the varieties themselves remain normal and irreducible).

D. Hodge Theory and Torelli Theorems (Theorems 5 & 6)

• Hodge Numbers: The paper computes the Hodge numbers $h^{p,q}$ of the crepant resolution $\tilde{X}$.
  • $h^{p,p}(\tilde{X}) = N_p$ (except when $n$ is even and $p=n/2$, where it is $2N_p$).
  • $h^{p,q} = 0$ unless $p=q$ or $p+q=n$.
  • Here, $N_p$ counts specific integer tuples related to the Sylvester sequence.
• Kodaira-Spencer Isomorphism: The map $T_{[X]}E_n \to H^1(T_{\tilde{X}})$ is an isomorphism.
• Local Torelli: Consequently, the period map is an immersion, establishing a local Torelli theorem for these families.

E. Generalization of Elliptic Surface Theory (Section 5)

The paper extends classical results from the theory of elliptic surfaces (Kodaira classification, Tate's algorithm, canonical class formulas) to higher dimensions. It defines a "minimal form" for the family and calculates the log canonical threshold of fibers, showing it depends on the valuation of the coefficients.

4. Significance

1. Explicit Compactifications: Prior to this work, explicit complete moduli spaces for high-dimensional Calabi-Yau varieties were largely unknown or non-constructive. This paper provides a concrete tower of weighted projective spaces serving as such compactifications.
2. Unification of Dimensions: It unifies the theory of elliptic curves ($n=1$) and K3 surfaces ($n=2$) with higher dimensions, showing that the moduli spaces for $n \geq 3$ share structural similarities (weighted projective spaces, Baily-Borel-like boundaries) with the lower-dimensional cases.
3. KSBA Stability: It demonstrates that these specific Calabi-Yau varieties fit naturally into the KSBA framework, providing a robust geometric invariant theory (GIT) interpretation.
4. Self-Mirror Duality: The paper notes that these varieties are self-mirror dual (generalizing Arnold's strange duality for $D_{2,3,7}$), linking the construction to mirror symmetry.
5. Computational Feasibility: By reducing the problem to weighted projective spaces and combinatorial counts ($N_p$), the paper makes the study of the Hodge theory and deformation theory of these high-dimensional varieties computationally tractable.

In summary, Alexeev constructs a "tower" of moduli spaces that generalizes the classical moduli of elliptic curves and K3 surfaces to arbitrary dimensions, proving they are complete, weighted projective spaces parameterizing Calabi-Yau varieties with controlled singularities, and establishing their fundamental geometric and Hodge-theoretic properties.