Some Remarks on Random Vectors and $O(n)$-Invariants

Alexander Kushkuley

2023-03-10

Abstract:Computations involving invariant random vectors are directly related to the theory of invariants (cf. e.g \cite{Weing_1}). Some simple observations along these lines are presented in this paper. We note in particular that sum of elements of the standard basis of $ O(n)$-invariants is equal to the expectation of a random Veronese tensor up to a known scalar multiplier.

Numerical Analysis,Probability,Representation Theory

What problem does this paper attempt to address?

The problems that this paper attempts to solve mainly focus on the relationship between random vectors and invariant theory, especially regarding the computation of \(O(n)\)-invariants. Specifically, the paper explores the following points:

1. **Relationship between random vectors and invariants**:
- The author studied the relationship between the expected value of an invariant random vector and its corresponding invariant projection in a real vector space \(V\) under the action of a compact Lie group \(G\). In particular, the author pointed out that the sum of the elements of the standard basis vector is equal to the expected value of the random Veronese tensor multiplied by a known scalar factor.

2. **Tensors and invariants**:
- The paper discussed the invariants in the tensor space \(V^{\otimes m}\), especially when \(m = 2k\), the properties of the \(O(V)\)-invariant \(I(P)\) defined by pairing. These invariants form a linear space in the tensor space, and their number and properties are determined by the first fundamental theorem (FFT) and the second fundamental theorem (SFT) of the invariant theory of the orthogonal group.

3. **Random tensors and Veronese surfaces**:
- The author introduced the equivariant Veronese mapping on the unit sphere and studied the expected value of the random Veronese tensor corresponding to the invariant random vector. Through these studies, the author reached an important conclusion about the expected value of the random Veronese tensor, that is, it is equal to the average value of the elements of the standard invariant basis divided by the average value of the elements of the Gram matrix.

4. **Application of Weingarten calculus**:
- The paper also explored the application of Weingarten calculus in computing the moments of Haar - distributed matrices. By transforming the moment problem into an invariant problem in the tensor space, the author provided a computational method, especially when dealing with high - order moments.

In summary, this paper aims to provide a new perspective and tool to understand and compute the invariants related to the orthogonal group \(O(n)\) by combining random vectors, tensor analysis, and invariant theory. This not only deepens our understanding of these mathematical structures but also provides a theoretical basis for further research in related fields.

In terms of formulas, the main formulas involved in the paper include:
- Invariant projection formula: \[P_G(v)=\int_G gv d\chi(g)\]
- Expected value of random Veronese tensor: \[E(x^{\otimes m = 2k})=\frac{(2k - 1)!!}{P(n,k)}A_m\equiv\mu_{k,n}A_m\]
- Inner product of Gram matrix elements: \[ \langle I(P),I(Q)\rangle=n^l\]

These formulas play a crucial role in the paper and help derive many important conclusions.