Neural Surrogate Filtering for Social Measure Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surrogate models for social simulations are accurate but time-consuming, while faster models lack sufficient accuracy, making it difficult to find optimal measures efficiently.
Innovation Solution
A system using a trained neural network-based surrogate model for pre-filtering measures, distinguishing between interpolation and extrapolation areas, and combining with detailed simulation for high-accuracy optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a social simulation is used to evaluate measures, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The evaluation process is segmented into two stages: first using a surrogate model for rapid preliminary evaluation of multiple measures, then using full social simulation only for the top candidate measures. This segmentation allows the system to achieve high measurement precision through simulation while reducing overall time loss by avoiding simulation of all measures.
Solution Approach 2:
A surrogate model acts as an intermediary between the full social simulation and the measure evaluation process. The surrogate model provides fast approximations that filter out inferior measures before simulation, thereby reducing the total calculation time while maintaining evaluation accuracy through subsequent simulation of promising candidates.
2Loss of time
If a surrogate model is used to substitute simulation, then loss of time is reduced, but measurement precision deteriorates
Solution Approach 1:
The system applies partial action by using the surrogate model for only the preliminary filtering stage rather than for all evaluations. The full social simulation is still applied to the top candidate measures, ensuring that measurement precision is maintained for the final selection while time loss is reduced through the use of the faster surrogate model for initial filtering.
3Measurement precision
If comprehensive evaluation of all measures is performed, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The surrogate model performs preliminary action by rapidly evaluating all candidate measures to identify the top performers before the more time-consuming social simulation is applied. This preliminary filtering action reduces the number of measures that require comprehensive simulation evaluation, thereby reducing overall optimization time while maintaining precision through simulation of the top candidates.
Data Source
AI summary
An information processing apparatus receives data for a plurality of social measures that is a consideration target, detects a prediction model constructed from a trained neural network for substituting a simulation related to social measures by using machine learning; performs, based on characteristics of the prediction model, a filtering processing on the data for the plurality of social measures, performs the simulation on data for one or more of social measures extracted by the filtering processing, specifies, based on a result of the simulation performed on the data for the one or more social measures, a first social measure that is an implementation target from among the plurality of social measures that is the consideration target, and outputs the first social measure.