[Paper Review] On the dimension of the Schur multiplier of nilpotent Lie algberas

Kavita Pradeep|arXiv (Cornell University)|May 9, 2017

Advanced Topics in Algebra10 references3 citations

TL;DR

This paper establishes a new upper bound for the dimension of the Schur multiplier of a finite-dimensional c-step nilpotent Lie algebra, improving upon prior bounds by incorporating the structure of lower central series and using representation-theoretic techniques involving tensor products and kernel dimensions of bilinear maps. The key result is a sharper bound that depends on the algebra's dimension, the derived algebra's dimension, and the nilpotency class.

ABSTRACT

We give a bound on the dimension of the Schur multiplier of a finite dimensional nilpotent Lie algebra which sharpens the earlier known bounds.

Motivation & Objective

• To refine existing upper bounds on the dimension of the Schur multiplier of finite-dimensional nilpotent Lie algebras.
• To address the gap in tightness of previous bounds that did not fully account for the nilpotency class and derived algebra structure.
• To provide a sharper, more precise bound that incorporates the dimension of kernels of natural bilinear maps arising from free presentations.
• To generalize and improve upon earlier results by Niroomand and Russo, as well as Yankosky and Hardy–Stitzinger, by introducing a more refined structural analysis.

Proposed method

• Utilizes a free presentation $ F/R $ of the Lie algebra $ L $, with $ F $ free and $ R $ the relator ideal.
• Applies the identity $ \dim(M(L)) = \dim(M(L/\gamma_c(L))) + \left(\dim(L/\gamma_2(L)) - 1\right)\dim(\gamma_c(L)) - \dim(\ker(\lambda_c)) $, derived from Eshrati et al.'s work.
• Defines bilinear maps $ \lambda_i: \frac{L}{\gamma_2(L)} \otimes \frac{\gamma_i(L)}{\gamma_{i+1}(L)} \to \frac{[F, \gamma_i(F)+R]}{[\gamma_{i+1}(F)+R, F]} $, whose kernels are analyzed to bound the Schur multiplier.
• Establishes linear independence of certain elements $ \Psi_i $ in $ \ker(\lambda_i) $ using a minimal generating set of size $ n - m $, where $ n = \dim(L) $, $ m = \dim(\gamma_2(L)) $.
• Uses induction on the nilpotency class $ c $, applying the dimension formula recursively to $ L/\gamma_i(L) $ for $ 2 \leq i \leq c $.
• Applies Corollary 3.2 to show $ \dim(\ker(\lambda_i)) \geq n - m - i $ for $ i \leq \min(n - m, c) $, leading to the final bound.

Experimental results

Research questions

• RQ1 Can the known upper bound for the Schur multiplier of a nilpotent Lie algebra be improved by incorporating the nilpotency class and derived algebra structure?
• RQ2 What is the precise contribution of the kernel dimensions of the bilinear maps $ \lambda_i $ to the Schur multiplier dimension?
• RQ3 How does the size of a minimal generating set influence the lower bound on $ \dim(\ker(\lambda_i)) $?
• RQ4 To what extent can the bound of Niroomand and Russo be sharpened using deeper structural analysis of the lower central series?
• RQ5 Is there a systematic way to refine the Schur multiplier bound by analyzing the tensor product structure of quotients in the lower central series?

Key findings

• The paper establishes a new upper bound: $ \dim(M(L)) \leq \frac{1}{2}(n - m - 1)(n + m) - \sum_{i=2}^{\min(n - m, c)} (n - m - i) $, where $ n = \dim(L) $, $ m = \dim(\gamma_2(L)) $, and $ c $ is the nilpotency class.
• This bound improves upon the earlier bound by Niroomand and Russo, which was $ \frac{1}{2}(n - m - 1)(n + m - 2) + 1 $, by including an additional sum term that accounts for higher-order commutators.
• The improvement arises from a lower bound on $ \dim(\ker(\lambda_i)) $, shown to be at least $ n - m - i $, derived from the linear independence of $ \Psi_i $-type elements in the kernel.
• The bound is sharp in the sense that it accounts for the full structure of the lower central series and the minimal generating set size, and it generalizes previous results by Yankosky and Hardy–Stitzinger.
• The result confirms that the Schur multiplier of a finite-dimensional nilpotent Lie algebra of dimension >1 is non-trivial, consistent with earlier results by Bosko, Stitzinger, and Riyahi–Salemkar.
• The paper suggests a further improvement is possible by subtracting $ \dim\left(\frac{Z(L)}{\gamma_2(L) \cap Z(L)}\right) \cdot \dim(\gamma_2(L)) $, indicating potential for even tighter bounds.