Stochastic Decision-Maker Modeling for Sequential Attribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In computing environments with multiple decision makers, existing technologies face challenges in identifying non-deterministic models where decisions are made sequentially and the actual decision makers are unknown, leading to difficulties in predicting behavior and understanding system impacts.
Innovation Solution
A cognitive system using machine learning operations like Kalman filters, particle filtering, and Monte Carlo Markov Chain techniques to identify non-deterministic models by analyzing sequences of decisions and their outcomes, allowing for the identification of individual decision makers and their transition matrices in a computing environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional modeling approaches are used in environments with multiple decision makers, then the system can handle sequential decisions, but it cannot accurately identify individual decision maker behavior or predict future decisions
Solution Approach 1:
The patent segments the collective decision-making process into individual decision maker models. By creating separate non-deterministic models for each decision maker and using algorithms to attribute decisions to specific individuals, the system can identify and predict individual behavior patterns while maintaining manageable model complexity through modular architecture.
Solution Approach 2:
The patent transforms the modeling approach by changing parameters from deterministic to non-deterministic models. This allows the system to capture the probabilistic nature of individual decision maker behavior, improving prediction accuracy by accounting for variability in decision patterns while using statistical parameters to manage complexity.
2Reliability
If the system attempts to identify individual decision makers from observed decisions, then prediction accuracy improves, but the computational complexity and difficulty of analysis increases
Solution Approach 1:
The patent implements feedback mechanisms where the system continuously observes decisions, updates non-deterministic models, and refines attribution algorithms. This iterative feedback process improves prediction reliability over time by learning from observed patterns while managing analytical difficulty through automated model updating and probabilistic attribution methods.
Solution Approach 2:
The patent replaces traditional mechanical attribution methods with computational algorithms including machine learning and probabilistic models. This substitution enables the system to handle the complexity of identifying individual decision makers from observed decisions by using automated computational approaches rather than manual analysis, improving reliability while managing difficulty through algorithmic processing.
Data Source
AI summary
Embodiments for identifying stochastic models representing the individual decision makers in a computing environment by a processor. One or more non-deterministic (stochastic, probabilistic) models may be identified according to a sequence of outcomes from decisions of each of a plurality of decision makers.