Synthesizing Action Sequences for Modifying Model Decisions
This addresses the need for interpretable and practical guidance in AI decision-making, particularly in high-stakes domains like finance, though it is an incremental advancement in applying synthesis techniques to model feedback.
The paper tackles the problem of generating actionable feedback for individuals to change unfavorable model decisions, such as loan denials, by synthesizing optimal sequences of actions using a combination of search-based program synthesis and test-time adversarial attacks.
When a model makes a consequential decision, e.g., denying someone a loan, it needs to additionally generate actionable, realistic feedback on what the person can do to favorably change the decision. We cast this problem through the lens of program synthesis, in which our goal is to synthesize an optimal (realistically cheapest or simplest) sequence of actions that if a person executes successfully can change their classification. We present a novel and general approach that combines search-based program synthesis and test-time adversarial attacks to construct action sequences over a domain-specific set of actions. We demonstrate the effectiveness of our approach on a number of deep neural networks.