# US 2024/0004953 A1 · Computing estimated closest correlation matrices

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2024/0004953 A1
- Document ID: US20240004953A1
- Status: Published application
- Kind: A1
- Application number: 17/810,249
- Filed: 2022-06-30
- Published: 2024-01-04
- Earliest claimed: 2022-06-30
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing, LLC
- Inventors: Bradley Curtis LACKEY; Andrew John MCGUINNESS
- Web page: https://patentgrep.com/patent/US20240004953A1

## CPC

- G06F 17/16: Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)}
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A computing system is provided, including one or more processors configured to receive a plurality of input matrices M. Each input matrix M may include a plurality of estimated input correlation coefficients. The one or more processors may be further configured to compute a respective plurality of estimated closest correlation matrices X0 for the plurality of input matrices M at a semidefinite program solver. Each estimated closest correlation matrix X0 may be a positive definite matrix. The one or more processors may be further configured to generate a training data set including at least the plurality of estimated closest correlation matrices X0. The one or more processors may be further configured to train a machine learning model using the training data set.

## Claims

### Claim 1

A computing system comprising:
one or more processors configured to:
receive a plurality of input matrices M, wherein each input matrix M includes a plurality of estimated input correlation coefficients;
compute a respective plurality of estimated closest correlation matrices X0 for the plurality of input matrices M at a semidefinite program solver, wherein each estimated closest correlation matrix X0 is a positive definite matrix;
generate a training data set including at least the plurality of estimated closest correlation matrices X0; and
train a machine learning model using the training data set.

### Claim 2 (depends on 1)

The computing system of claim 1, wherein the one or more processors are configured to estimate each estimated closest correlation matrix X0 to be the positive definite matrix closest to the corresponding input matrix M according to a least-squares distance measure.

### Claim 3 (depends on 2)

The computing system of claim 2, wherein, at the semidefinite program solver, the one or more processors are configured to compute the respective estimated closest correlation matrix X0 for each of the input matrices M at least in part by estimating a candidate solution matrix X that minimizes tr(MX).

### Claim 4 (depends on 3)

The computing system of claim 3, wherein, at the semidefinite program solver, the one or more processors are further configured to:
receive a smallest eigenvalue λ of the estimated closest correlation matrix X0; and
compute the estimated closest correlation matrix X0 under a constraint that tr(AiX)=1−λ, where Ai is a square matrix in which each element is equal to 0 except a 1 located along a main diagonal of Ai in an ith row.

### Claim 5 (depends on 1)

The computing system of claim 1, wherein the one or more processors are configured to receive the plurality of input matrices M via user input at a graphical user interface (GUI).

### Claim 6 (depends on 1)

The computing system of claim 1, wherein the one or more processors are configured to compute the estimated input correlation coefficients based at least in part on a copula.

### Claim 7 (depends on 1)

The computing system of claim 1, wherein the one or more processors are configured to compute a plurality of estimated input correlation coefficients from empirical correlation data.

### Claim 8 (depends on 1)

The computing system of claim 1, wherein the training data set further includes a plurality of marginal distributions.

### Claim 9 (depends on 1)

The computing system of claim 1, wherein the plurality of marginal distributions are financial risk distributions.

### Claim 10 (depends on 1)

The computing system of claim 1, wherein the plurality of marginal distributions are energy source availability distributions.

### Claim 11

A method for use with a computing system, the method comprising:
receiving a plurality of input matrices M, wherein each input matrix M includes a plurality of estimated input correlation coefficients;
computing a respective plurality of estimated closest correlation matrices X0 for the plurality of input matrices M at a semidefinite program solver, wherein each estimated closest correlation matrix X0 is a positive definite matrix;
generating a training data set including at least the plurality of estimated closest correlation matrices X0; and
training a machine learning model using the training data set.

### Claim 12 (depends on 11)

The method of claim 11, wherein each estimated closest correlation matrix X0 is estimated to be the positive definite matrix closest to the corresponding input matrix M according to a least-squares distance measure.

### Claim 13 (depends on 12)

The method of claim 12, wherein, at the semidefinite program solver, the respective estimated closest correlation matrix X0 is computed for each of the input matrices M at least in part by estimating a candidate solution matrix X that minimizes tr(MX).

### Claim 14 (depends on 13)

The method of claim 13, further comprising, at the semidefinite program solver:
receiving a smallest eigenvalue λ of the estimated closest correlation matrix X0; and
computing the estimated closest correlation matrix X0 under a constraint that tr(AiX)=1−λ, where Ai is a square matrix in which each element is equal to 0 except a 1 located along a main diagonal of Ai in an ith row.

### Claim 15 (depends on 11)

The method of claim 11, wherein the plurality of input matrices M are received via user input at a graphical user interface (GUI).

### Claim 16 (depends on 11)

The method of claim 11, further comprising computing the estimated input correlation coefficients based at least in part on a copula.

### Claim 17 (depends on 11)

The method of claim 11, further comprising computing a plurality of estimated input correlation coefficients from empirical correlation data.

### Claim 18 (depends on 11)

The method of claim 11, wherein the training data set further includes a plurality of marginal distributions.

### Claim 19 (depends on 11)

The method of claim 11, wherein the plurality of marginal distributions are financial risk distributions or energy source availability distributions.

### Claim 20

A computing system comprising:
one or more processors configured to:
at a graphical user interface (GUI), receive:
a user input specifying an input matrix M that includes a plurality of estimated input correlation coefficients; and
a smallest eigenvalue λ for an estimated closest correlation matrix X0;
compute the estimated closest correlation matrix X0 for the input matrix M at a semidefinite program solver, wherein the estimated closest correlation matrix X0 is a positive definite matrix and has the smallest eigenvalue λ; and
output a graphical representation of the estimated closest correlation matrix X0 for display at the GUI.
