Abstract:In this paper, we study the product of two simple modules over KLR algebras using the quiver Grassmannians for Dynkin quivers. More precisely, we establish a bridge between the Induction functor on the category of modules of KLR algebras and the irreducible components of quiver Grassmannians for Dynkin quivers via a sort of extension varieties, which is an analogue of the extension group in Hall algebras. As a result, we give a necessary condition when the product of two simple modules over a KLR algebra is simple using the set of irreducible components of quiver Grassmannians. In particular, in some special cases, we provide a proof for the conjecture recently proposed by Lapid and Minguez.
What problem does this paper attempt to address?
The problems that this paper attempts to solve mainly focus on two aspects:
1. **Study the product of two simple modules over the KLR algebra**: Specifically, the author tries to build a bridge between the induction functor on the category of KLR - algebra modules and the irreducible components of the matroid Grassmannian through the matroid Grassmannian of the Dynkin quiver. This connection is achieved through some extended cluster, which is similar to the extended group in the Hall algebra.
2. **Give the necessary conditions for the product of simple modules to be a simple module**: Based on the set of irreducible components of the matroid Grassmannian, the author provides a necessary condition. That is, in some special cases, when the product of two simple modules over the KLR algebra is also a simple module, this condition holds. In particular, the author provides a partial proof for the conjecture recently proposed by Lapid and Minguez.
### Specific problem description
- **Problem background**:
- The paper considers the semi - part of the simple Lie algebra \( \mathfrak{g} \) and the corresponding quantum group \( U_q(\mathfrak{n}) \).
- By geometrizing \( U_q(\mathfrak{n}) \), Lusztig introduced the dual canonical basis \( B^* \).
- Studying the product of dual canonical basis elements is an important problem. For example, Berenstein and Zelevinsky proposed a conjecture, but Leclerc showed that this conjecture does not hold in general and proposed the concept of real dual canonical basis elements.
- **The connection between the KLR algebra and the matroid Grassmannian**:
- By using the module category of the quiver Hecke algebra \( R_Q \) to classify \( U_q(\mathfrak{n}) \) (or \( A_q(\mathfrak{n}) \)), the dual canonical basis elements can be regarded as simple modules over \( R_Q \).
- Kang, Kashiwara, Kim and Oh used the R - matrix to prove Leclerc's conjecture.
- Lapid and Minguez proposed a conjecture about irreducible components, which links the product of two dual canonical basis elements to the representation of the pre - projective algebra.
- **Main problem**:
- Given two positive - dimensional vectors \( \alpha \) and \( \beta \), and a pair of Kostant partitions \( (\mu, \nu) \in \text{KP}(\alpha) \times \text{KP}(\beta) \), describe the conditions such that the induced module \( L(\mu) \circ L(\nu) \) is still a simple module.
- This problem was proposed by Kleshchev and Ram in [KR11, Problem 7.6], but their description uses the vertex set of the quiver instead of Kostant partitions.
### Solution
- **Build a bridge**:
- The author establishes the connection between the representation of the quiver \( Q \) and the \( R_Q \) - module through the irreducible components of the matroid Grassmannian.
- In particular, the author gives the sufficient conditions for \( q^n L(\mu \star \nu) \cong \text{soc} L(\mu) \circ L(\nu) \) to hold, where \( n \) is some integer.
- **Main results**:
- If \( L(\mu) \circ L(\nu) \) is a simple module, then \( (\mu, \nu) \) is a support pair.
- For any non - trivial extension \( \lambda \in \text{ext}(\mu, \nu) \), if \( (\mu, \nu) \) is the universal pair of \( \lambda \), the following condition is satisfied:
- \( \dim \text{Hom}_Q(M_\nu, M_\mu \oplus M_\nu) \)