Lifting derived equivalences of abelian surfaces to generalized Kummer varieties

This paper establishes a $G$-equivariant analogue of Orlov's short exact sequence for derived categories of abelian varieties and utilizes this framework, along with splitting propositions, to lift derived equivalences of abelian surfaces to generalized Kummer varieties.

The Gist

Imagine you are an architect working with two very different types of buildings: Abelian Surfaces and Generalized Kummer Varieties.

In the world of mathematics (specifically algebraic geometry), these aren't just buildings; they are complex shapes with hidden symmetries. Mathematicians study them not just by looking at their shape, but by studying their "blueprints"—a mathematical structure called a Derived Category. Think of the Derived Category as the ultimate instruction manual for how to build, deconstruct, and rearrange the pieces of these shapes.

Sometimes, two different shapes have the exact same instruction manual. If you can swap the pieces of Shape A with the pieces of Shape B without breaking the rules, we say they are Derived Equivalent. It's like realizing that a Lego castle and a Lego spaceship are built from the exact same set of bricks, just arranged differently.

The Problem: The "Hard Mode" of Geometry

The author, Yuxuan Yang, tackles a specific challenge:

1. Shape A is an Abelian Surface (think of it as a fancy, multi-dimensional donut).
2. Shape B is a Generalized Kummer Variety (a much more complex, twisted shape created by taking Shape A, making $n$ copies of it, and gluing them together in a specific, symmetrical way).

Mathematicians already knew how to find the "instruction manuals" (equivalences) for the simple donuts (Abelian Surfaces). But finding the manuals for the complex, twisted Kummer shapes was like trying to solve a Rubik's Cube while blindfolded. It was incredibly difficult.

The Solution: The "Equivariant" Elevator

Yang's paper introduces a clever method to solve this. He uses a concept called G-Equivariance.

The Analogy of the Dance Floor:
Imagine the Abelian Surface is a dance floor.

• The "G" Group: Imagine a group of dancers (a finite subgroup) who have a specific rule: they must move in perfect sync. If one dancer moves left, they all move left.
• The "Equivariant" Category: This is the study of the dance floor only considering moves that respect this group's rules. It's like studying the dance floor while wearing glasses that only let you see the synchronized moves.

Yang discovered a secret "elevator" (a mathematical sequence) that connects the synchronized dance moves of the simple floor (Abelian Surface) to the complex, twisted floor (Kummer Variety).

How the "Lifting" Works

The core idea of the paper is Lifting.

1. Start Simple: You find a cool, symmetrical move on the simple Abelian Surface (a "Derived Equivalence").
2. Check the Rules: You check if this move respects the "G" group's synchronization rules.
3. The Lift: If it does, Yang's method allows you to "lift" this move up to the complex Kummer Variety.

It's like finding a simple pattern in a small tile and realizing that this exact pattern can be scaled up to decorate a massive, intricate cathedral ceiling. You don't have to design the cathedral from scratch; you just take the proven pattern from the tile and "lift" it to the ceiling.

The "Splitting" Trick

One of the most beautiful parts of the paper is the Splitting Theorem.

When you lift a move from the simple surface to the complex Kummer variety, the resulting move on the Kummer variety turns out to be a combination of two simpler things:

1. A move that happens only on the complex Kummer shape.
2. A move that happens only on the original simple Abelian surface.

The Metaphor:
Imagine you have a complex machine (the Kummer Variety) that is actually just a simple machine (the Abelian Surface) sitting next to a new, fancy attachment.
Yang proves that if you know how to operate the simple machine, you can automatically figure out how to operate the whole complex machine. The complex machine's behavior "splits" into the behavior of the simple machine plus a specific, predictable adjustment.

Why Does This Matter?

• Solving the Unsolvable: Before this, finding symmetries for these complex Kummer shapes was a nightmare. This paper provides a "cheat code." If you can solve it for the simple donut, you can solve it for the complex Kummer shape.
• K3 Surfaces: The paper specifically looks at a special case where the Kummer shape becomes a K3 Surface (a very famous and important shape in physics and math, related to string theory). Yang shows exactly which symmetries of the simple donut can be lifted to these K3 surfaces and which ones cannot.
• The "Spin" Connection: The paper uses some heavy-duty math involving "Spin groups" (related to the physics of electrons) to prove that this lifting works. It's like using the laws of quantum physics to prove a theorem about architecture.

Summary

In plain English, Yuxuan Yang wrote a guidebook that says:

"If you want to understand the complex, twisted symmetries of these special geometric shapes (Kummer Varieties), don't try to figure them out from scratch. Instead, look at the simpler shapes (Abelian Surfaces) they are built from. If you find a symmetry in the simple shape that follows the group rules, you can 'lift' it up to the complex shape. Furthermore, the complex shape's symmetry is just the simple shape's symmetry plus a predictable extra piece."

This bridges the gap between the known and the unknown, turning a mountain of difficult math into a manageable staircase.

Technical Summary

Here is a detailed technical summary of the paper "Lifting Derived Equivalences of Abelian Surfaces to Generalized Kummer Varieties" by Yuxuan Yang.

1. Problem Statement

The paper addresses the problem of constructing and classifying derived equivalences (Fourier-Mukai equivalences) between generalized Kummer varieties, which are hyper-Kähler manifolds constructed from abelian surfaces.

While derived equivalences for K3 surfaces and their Hilbert schemes are well-understood (notably via the work of Ploog, who lifted equivalences from K3 surfaces to their Hilbert schemes $S^{[n]}$), the situation for generalized Kummer varieties is more complex. These varieties, denoted $Kum_{n-1}(A)$, are defined as the fibers of the summation map $\Sigma: A^{[n]} \to A$ (or related constructions involving the kernel of the summation map on $A^n$).

The central challenge is:

1. How to lift derived equivalences (or autoequivalences) from an abelian surface $A$ to the generalized Kummer variety $Kum_{n-1}(A)$.
2. How to characterize the group of derived autoequivalences of $Kum_{n-1}(A)$ in terms of the symplectic geometry and cohomology of the underlying abelian surface.

2. Methodology

The author employs a multi-step methodology combining category theory, representation theory, and algebraic geometry:

• G-Equivariant Categories: The paper utilizes the theory of $G$-equivariant derived categories, $D^b_G(A)$, where $G$ is a finite subgroup of the dual abelian variety $\hat{A} \cong \text{Pic}^0(A)$. The author establishes that $D^b_G(A)$ is equivalent to the derived category of another abelian variety $B$ (where $G = \ker(\hat{B} \to \hat{A})$).
• Orlov's Short Exact Sequence (G-Equivariant Version): The core theoretical tool is a generalization of Orlov's short exact sequence for derived equivalences of abelian varieties. The author constructs a $G$-equivariant version of this sequence, relating the group of $G$-autoequivalences of $D^b_G(A)$ to a specific subgroup of Hodge isometries on the cohomology lattice.
• Bridgeland-King-Reid (BKR) Correspondence: The paper leverages the BKR theorem, which establishes an equivalence between the derived category of the $S_n$-equivariant category of $A^n$ (or related spaces) and the derived category of the Hilbert scheme or generalized Kummer variety. Specifically, $D^b_{S_n}(NA \times A) \simeq D^b(Kum_{n-1}(A) \times A)$.
• Lifting and Splitting: The author defines functors to "lift" equivalences from the abelian surface level to the equivariant level, and then to the generalized Kummer level. A crucial step involves proving a splitting theorem, showing that lifted equivalences on the product space $Kum_{n-1}(A) \times A$ decompose uniquely into a product of an equivalence on $Kum_{n-1}(A)$ and an equivalence on $A$.
• Spin Representations and Triality: For the specific case of Kummer K3 surfaces ($n=2$), the author utilizes the triality of $\text{Spin}(8)$ to relate the symplectic maps on the cohomology of the abelian surface to the Hodge isometries on the Mukai lattice of the K3 surface.

3. Key Contributions and Results

A. The G-Equivariant Orlov Sequence (Section 1)

The author proves a $G$-equivariant analogue of Orlov's short exact sequence.

• Theorem 1.17: For an abelian variety $A$ and a finite subgroup $G \leq \text{Pic}^0(A)$, there is a short exact sequence:
  $$0 \to \text{Alb}(B) \times \hat{A} \times \mathbb{Z} \to G\text{-Aut}(D^b_G(A)) \xrightarrow{G-\rho_A} G\text{-SO}^+_{\text{Hdg}}(V_B) \to 0$$
  Here, $G\text{-SO}^+_{\text{Hdg}}(V_B)$ is a subgroup of Hodge isometries preserving specific lattices related to the quotient map $q: B \to A$.
• Significance: This result characterizes exactly which derived equivalences of the underlying abelian variety can be lifted to $G$-equivariant equivalences. It establishes a surjective map from the group of $G$-autoequivalences to a specific subgroup of symplectic maps.

B. Lifting to Generalized Kummer Varieties (Section 2)

The paper constructs a mechanism to lift equivalences from $D^b(A)$ to $D^b(Kum_{n-1}(A))$.

• Construction: By using the map $q: NA \times A \to A^n$ and the BKR equivalence, the author lifts $G$-autoequivalences of $D^b_G(A)$ to autoequivalences of $D^b(Kum_{n-1}(A) \times A)$.
• The Splitting Theorem (Theorem 2.5 & Corollary 2.6):
  For any $G$-equivariant derived equivalence $(f, \sigma)$, the induced autoequivalence on $D^b(Kum_{n-1}(A) \times A)$ splits uniquely:
  $$\tilde{\lambda}_q(\tilde{\delta}_A(f, \sigma)) = \Phi_{(f,\sigma)} \times \Psi_{(f,\sigma)}$$
  where $\Phi_{(f,\sigma)} \in \text{Aut}(D^b(Kum_{n-1}(A)))$ and $\Psi_{(f,\sigma)} \in \text{Aut}(D^b(A))$.
• Implication: This allows the construction of a series of derived autoequivalences for generalized Kummer varieties directly from those of the abelian surface.

C. Characterization via Symplectic Maps (Theorem 2.12 & Corollary 2.13)

The author provides a concrete criterion for identifying which derived equivalences of generalized Kummer varieties arise from this lifting process.

• A derived equivalence $\Phi$ of $Kum_{n-1}(A)$ is of the "lifted" form if and only if its induced symplectic map on cohomology, $\gamma_{NA}(\Phi)$, has a specific block-matrix structure derived from a symplectic map $g \in \text{Sp}(A)$ satisfying a lattice condition:
  $$(g_2)_* \left( \frac{1}{n} H^1(\hat{A}, \mathbb{Z}) \right) \subset n H^1(A, \mathbb{Z})$$
  This identifies the image of the lifting map as a subgroup of finite index ($n^2$) within the full group of derived autoequivalences.

D. Application to Kummer K3 Surfaces (Section 2.5)

For $n=2$, $Kum_1(A)$ is a Kummer K3 surface.

• The paper investigates whether the lifting method covers all derived autoequivalences of the Kummer K3 surface.
• Using the triality of $\text{Spin}(8)$, the author translates the condition on the symplectic map of the abelian surface into a condition on the even cohomology of the K3 surface.
• Result: The set of lifted equivalences forms a proper subgroup of the full group of derived autoequivalences of the Kummer K3 surface. This implies there are "exotic" autoequivalences of Kummer K3 surfaces that do not arise from the abelian surface via this specific lifting mechanism.

4. Significance

• Structural Insight: The paper provides a rigorous framework for understanding the relationship between the derived categories of abelian varieties and their associated hyper-Kähler varieties (generalized Kummer varieties).
• Generalization of Ploog's Work: It extends the known results for K3 surfaces and Hilbert schemes to the more complex setting of generalized Kummer varieties, where the group actions and lattice structures are more intricate.
• Classification Tool: The explicit description of the lifted equivalences via symplectic maps and Hodge isometries offers a powerful tool for classifying derived equivalences in this context.
• Discovery of Exotic Equivalences: By showing that the lifting process does not capture all autoequivalences for Kummer K3 surfaces, the paper highlights the existence of derived symmetries that are intrinsic to the K3 geometry and not merely inherited from the underlying abelian surface. This deepens the understanding of the "derived Torelli" problem for these varieties.

In summary, Yang's work successfully bridges the gap between equivariant derived categories of abelian varieties and the geometry of generalized Kummer varieties, providing both a constructive method for generating equivalences and a precise characterization of their limitations.