Distance-Restricted Explanations: Theoretical Underpinnings & Efficient Implementation

2024-05-14

International Conference on Principles of Knowledge Representation and Reasoning

AI Summary

To address the real-time and correctness challenges of logically verifiable eXplainable AI (XAI) explanations in high-risk scenarios, this paper proposes the Distance-Restricted Explanation (DRE) framework, which rigorously guarantees both logical completeness and correctness of explanations within an input neighborhood—establishing, for the first time, a formal theoretical proof of completeness. Methodologically, DRE integrates Satisfiability Modulo Theories (SMT) solving, abstract interpretation, local robustness analysis, and incremental Boolean satisfiability checking, augmented with symbolic reasoning-based pruning and caching mechanisms to overcome the scalability bottlenecks of conventional logical explainers on large-scale inputs. Experiments demonstrate up to 100× inference acceleration on medium-complexity ML models, enabling real-time explanation generation for inputs of size ∼1,000, with near-zero error rates. DRE effectively bridges the gap between logical XAI and adversarial robustness.

Technology Category

• Humans and AI: Explainable AI (XAI) for Human Understanding
• Knowledge Representation and Reasoning: Description Logics
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

The uses of machine learning (ML) have snowballed in recent years. In many cases, ML models are highly complex, and their operation is beyond the understanding of human decision-makers. Nevertheless, some uses of ML models involve high-stakes and safety-critical applications. Explainable artificial intelligence (XAI) aims to help human decision-makers in understanding the operation of such complex ML models, thus eliciting trust in their operation. Unfortunately, the majority of past XAI work is based on informal approaches, that offer no guarantees of rigor. Unsurprisingly, there exists comprehensive experimental and theoretical evidence confirming that informal methods of XAI can provide human-decision makers with erroneous information. Logic-based XAI represents a rigorous approach to explainability; it is model-based and offers the strongest guarantees of rigor of computed explanations. However, a well-known drawback of logic-based XAI is the complexity of logic reasoning, especially for highly complex ML models. Recent work proposed distance-restricted explanations, i.e. explanations that are rigorous provided the distance to a given input is small enough. Distance-restricted explainability is tightly related with adversarial robustness, and it has been shown to scale for moderately complex ML models, but the number of inputs still represents a key limiting factor. This paper investigates novel algorithms for scaling up the performance of logic-based explainers when computing and enumerating ML model explanations with a large number of inputs.

Problem

• Explainable AI (XAI)
• Large-scale Input Data
• Computational Efficiency

Innovation

• Efficient Algorithm
• Explainable AI (XAI)
• Data-intensive Models