ExplainabilityTransparencyAccountabilitycs.LG
Partition-Guided Distance Saliency: Bridging Decision and Objective Spaces in Many-Objective Optimization
Cláudio Lúcio do Val Lopes, Flávio Vinícius Cruzeiro Martins, Elizabeth Fialho Wanner
June 29, 2026
arXiv.org
Key Takeaways
-
The "cognitive drought" problem: As optimization problems involve more objectives (goals), decision-makers struggle to understand trade-offs and make informed choices because traditional visualization methods break down—PGDS addresses this by automatically identifying meaningful target regions rather than requiring experts to manually specify them.
-
Bridging the explanation gap: The framework connects what changes in input variables (decision space) to what happens in outcomes (objective space) through geometric intuition, making it easier for non-technical stakeholders to understand which factors actually drive improvements toward their preferred solutions.
-
Automated feature categorization: Rather than treating all input variables equally, PGDS automatically identifies "Drivers" (variables that help reach desired outcomes) and "Blockers" (variables that constrain progress), enabling decision-makers to focus on the factors that matter most for their specific goals.
-
Practical automation reduces expertise barriers: By automatically partitioning the objective landscape and identifying improvement targets, the approach reduces the need for deep technical knowledge about the optimization problem, making complex multi-objective decisions more accessible to domain experts without AI backgrounds.
Abstract
Explainability in Many-Objective Optimization (MaO) is currently hindered by the escalating complexity of the Pareto front, which renders the relationship between high-dimensional decision variables and objective outcomes increasingly opaque. As the number of objectives exceeds the limits of traditional visualization, decision-makers encounter a ``cognitive drought'' in identifying relevant trade-offs or specifying target regions without a priori knowledge. To bridge this interpretability gap, we introduce the {Partition-Guided Distance Saliency (PGDS)} framework, a novel XAI approach designed for continuous optimization landscapes. Our framework automates the explanation process through a three-stage pipeline that prioritizes geometric intuition over abstract rules. First, we employ a surrogate model that learns how geometric distances in the decision space map to proximity in the objective space. Second, to address the difficulty of manual target selection in high dimensions, the framework automatically partitions the objective landscape into distinct regions and identifies local ``Dominating Points'' to serve as automated targets for improvement. Third, we quantify how sensitive a solution's position is to each decision variable by measuring the distance shifts induced by perturbations to each variable. This allows PGDS to categorize features as either ``Drivers'' which facilitate convergence toward preferred regions, or ``Blockers'' which represent geometric constraints hindering further progress. Validation on 10-objective benchmarks and a physics-informed engineering problem (Welded Beam) demonstrates that PGDS provides differentiated, actionable insights that traditional visualization and rule-based XAI methods fail to provide.