Multi-Step Recourse Paths for Feasible ML Decision Reversal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine learning models generate single-step recourse paths that are often unrealistic or infeasible, failing to provide actionable steps for individuals to achieve positive outcomes, and lack transparency in decision-making processes, particularly in high-stakes applications like credit line applications, violating regulatory mandates.
Innovation Solution
A method and system to augment machine learning models with feasible transitions by calculating an optimal distance function and generating a multi-step recourse path using an augmentation algorithm, ensuring each step is within a predetermined threshold, thereby providing realistic and actionable steps to achieve a positive outcome.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If single-step recourse paths are generated by machine learning models, then the decision-making process is simplified and faster, but the recourse paths become unrealistic and infeasible for actual implementation
Solution Approach 1:
The patent divides the single-step recourse path into multiple sequential steps, where each step represents a feasible transition between data points. By segmenting the overall recourse into intermediate steps with distance constraints, the system maintains computational efficiency while ensuring each individual step is realistic and actionable for the user.
2Reliability
If multi-step recourse paths with feasible transitions are generated, then the realism and actionability of recourse paths improve, but the computational complexity and processing time increase
Solution Approach 1:
The system pre-calculates and stores transition labels between data points during an offline phase, creating a lookup table of feasible transitions. During online recourse generation, the system only needs to query these pre-computed transitions and select appropriate steps, avoiding expensive real-time calculations and significantly reducing computational complexity while maintaining multi-step feasibility.
3Loss of information
If machine learning models provide decision explanations, then transparency and regulatory compliance improve, but the explanations may lack actionable guidance for individuals to change outcomes
Solution Approach 1:
The system generates recourse paths that provide feedback to users on what specific changes to their profile would lead to positive outcomes. Each step in the recourse path shows a feasible transition and its impact on the decision, creating an interactive feedback loop that guides users through actionable changes while maintaining transparency about the decision-making process and its reversibility.
Data Source
AI summary
A method and a system for determining a recourse path with respect to a decision that is associated with a positive outcome and a negative outcome are provided. The method includes: receiving a dataset including a data point representing an entity; determining, via a trained model, which data points from the dataset reach a positive outcome and which data points reach a negative outcome based on a distance threshold; determining transition labels from historical data; calculating an optimal distance function and an optimal threshold value for the dataset based on the transition labels; generating an augmentation algorithm based on the optimal distance function and the optimal threshold value; and generating a first recourse path for the first entity to reach the positive outcome by applying the augmentation algorithm to insert a second data point into the at least one dataset.