# US 2024/0005058 A1 · Stratifying conditional cumulative distribution models

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

- Number: US 2024/0005058 A1
- Document ID: US20240005058A1
- Status: Published application
- Kind: A1
- Application number: 17/810,276
- Filed: 2022-06-30
- Published: 2024-01-04
- Earliest claimed: 2022-06-30
- Assignees: Willis Group Limited
- Applicants: Willis Group Limited
- Inventors: Andrew John MCGUINNESS; Bradley Curtis LACKEY
- Web page: https://patentgrep.com/patent/US20240005058A1

## CPC

- G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30)
- G06F 2111/08: Probabilistic or stochastic CAD
- G06F 2111/10: Numerical modelling
- G06Q 40/08: Insurance
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

A computing system comprises a processor configured to receive, a plurality of simulations, a discrete distribution function, and one or more cumulative distribution models. One or more conditional cumulative distribution models are generated based at least in part on the discrete distribution function and the one or more cumulative distribution models. A range of the one or more cumulative distribution models and one or more conditional cumulative distribution models is stratified, and a sum of discrepancy scores is computed for the plurality of simulations. In each of one or more resampling iterations, one or more simulations are replaced with one or more resampled simulations based on a policy. An updated sum of discrepancy scores is generated for the plurality of simulations with the one or more simulations replaced by the one or more resampled simulations. The plurality of simulations are output subsequent to performing the one or more resampling iterations.

## Claims

### Claim 1

A computing system, comprising:
a processor configured to,
receive,
a plurality of simulations, wherein each simulation includes a plurality of simulation results,
a discrete distribution function, and
one or more cumulative distribution models;
generate one or more conditional cumulative distribution models based at least in part on the discrete distribution function and the one or more cumulative distribution models;
stratify a range of the one or more cumulative distribution models and one or more conditional cumulative distribution models into a number of strata;
compute a sum of discrepancy scores for the plurality of simulations based at least in part on the cumulative distribution models and conditional cumulative distribution models;
in each of one or more resampling iterations, until the sum of one or more respective discrepancy scores is determined to meet an optimization threshold:
generate one or more resampled simulations based at least in part on the one or more cumulative distribution models;
replace one or more simulations of the plurality of simulations with the one or more resampled simulations based on a policy, and
generate an updated sum of discrepancy scores for the plurality of simulations with the one or more simulations replaced by the one or more resampled simulations; and
output the plurality of simulations subsequent to performing the one or more resampling iterations.

### Claim 2 (depends on 1)

The computing system of claim 1, wherein the discrete distribution function is computed based upon the quantity of simulation results contained in each simulation.

### Claim 3 (depends on 2)

The computing system of claim 2, wherein the discrete distribution function forms a Poisson distribution.

### Claim 4 (depends on 1)

The computing system of claim 1, wherein the conditional cumulative distribution model is conditional upon a predetermined quantity of simulation results.

### Claim 5 (depends on 1)

The computing system of claim 1, wherein the processor is configured to approximate the conditional cumulative distribution model using a Fourier transform, a Monte Carlo method, or a Fourier transform based upon Monte Carlo simulation data.

### Claim 6 (depends on 1)

The computing system of claim 1, wherein the resampled simulation result is derived from a precomputed simulation.

### Claim 7 (depends on 1)

The computing system of claim 1, wherein the policy to replace one or more simulation results with one or more sampled simulation results is implemented at a Markov Chain Monte Carlo agent.

### Claim 8 (depends on 1)

The computing system of claim 1, wherein the resampling iterations are defined by a quantum-inspired algorithm.

### Claim 9 (depends on 7)

The computing system of claim 7, wherein the quantum-inspired algorithm is a Quantum Monte Carlo, Substochastic Monte Carlo, Population Annealing, or Parallel Tempering algorithm.

### Claim 10 (depends on 1)

The computing system of claim 1, wherein the processor is configured to apply the policy to minimize the one or more respective discrepancy scores.

### Claim 11

At a computing system, a method comprising:
receiving,
a plurality of simulations, wherein each simulation includes a plurality of simulation results,
a discrete distribution function, and
one or more cumulative distribution models;
generating one or more conditional cumulative distribution models based at least in part on the discrete distribution function and the one or more cumulative distribution models;
stratifying a range of the one or more cumulative distribution models and one or more conditional cumulative distribution models into a number of strata;
computing a sum of discrepancy scores for the plurality of simulations based at least in part on the cumulative distribution models and conditional cumulative distribution models;
in each of one or more resampling iterations, until the sum of one or more respective discrepancy scores is determined to meet an optimization threshold,
generating one or more resampled simulations based at least in part on the one or more cumulative distribution models,
replacing one or more simulations of the plurality of simulations with the one or more resampled simulations based on a policy, and
generating an updated sum of discrepancy scores for the plurality of simulations with the one or more simulations replaced by the one or more resampled simulations; and
outputting the plurality of simulations subsequent to performing the one or more resampling iterations.

### Claim 12 (depends on 11)

The method of claim 11, further comprising computing the discrete distribution function based upon the quantity of simulation results contained in each simulation.

### Claim 13 (depends on 12)

The method of claim 12, wherein the discrete distribution function forms a Poisson distribution.

### Claim 14 (depends on 11)

The method of claim 11, wherein the conditional cumulative distribution model is conditional upon a predetermined quantity of simulation results.

### Claim 15 (depends on 11)

The method of claim 11, further comprising approximating the conditional cumulative distribution model using a Fourier transform, a Monte Carlo method, or a Fourier transform based upon Monte Carlo simulation data.

### Claim 16 (depends on 11)

The method of claim 11, further comprising deriving the resampled simulation result from a precomputed simulation.

### Claim 17 (depends on 11)

The method of claim 11, further comprising implementing the policy at a Markov Chain Monte Carlo agent.

### Claim 18 (depends on 11)

The method of claim 11, wherein the resampling iterations are defined by a quantum-inspired algorithm.

### Claim 19 (depends on 18)

The method of claim 18, wherein the quantum-inspired algorithm includes a Quantum Monte Carlo, Substochastic Monte Carlo, Population Annealing, or Parallel Tempering algorithm.

### Claim 20

A computing system, comprising:
a processor configured to
receive,
a plurality of simulations, wherein each simulation includes a plurality of simulation results,
a discrete distribution function, and
one or more cumulative distribution models;
generate one or more conditional cumulative distribution models based at least in part on the discrete distribution function and the one or more cumulative distribution models;
stratify a range of the one or more cumulative distribution models and one or more conditional cumulative distribution models into a number of strata;
compute a sum of discrepancy scores for the plurality of simulations based at least in part on the cumulative distribution models and conditional cumulative distribution models;
in each of one or more resampling iterations defined by a quantum-inspired algorithm, until the sum of one or more respective discrepancy scores is determined to meet an optimization threshold:
generate one or more resampled simulations based at least in part on the one or more cumulative distribution models;
replace one or more simulations of the plurality of simulations with the one or more resampled simulations based on a policy configured to minimize the one or more respective discrepancy scores, and
generate an updated sum of discrepancy scores for the plurality of simulations with the one or more simulations replaced by the one or more resampled simulations; and
output the plurality of simulations subsequent to performing the one or more resampling iterations.
