MASCOTS: Model-Agnostic Symbolic COunterfactual explanations for Time Series

2025-03-28

AI Summary

Time-series counterfactual explanations face challenges due to strong temporal dependencies, high dimensionality, and lack of human-readable representations. Method: We propose a model-agnostic symbolic counterfactual generation framework that innovatively integrates Bag-of-Receptive-Fields with Symbolic Aggregate approXimation (SAX) to construct an interpretable symbolic feature space. Within this space, we perform gradient- and sampling-free optimization subject to symbolic distance constraints and feasibility regularization, yielding minimal yet semantically plausible input perturbations. The method natively supports multivariate time series. Contributions/Results: Our approach significantly improves explanation sparsity (↑3.8×), semantic readability (↑42%), and expressiveness in visual and natural language formats. Evaluated across diverse benchmark datasets, it achieves state-of-the-art performance in effectiveness, proximity, and plausibility.

Technology Category

• Cognitive Modeling & Cognitive Systems: Symbolic Representations
• Computer Vision: Visual Reasoning & Symbolic Representations
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

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

Abstract

Counterfactual explanations provide an intuitive way to understand model decisions by identifying minimal changes required to alter an outcome. However, applying counterfactual methods to time series models remains challenging due to temporal dependencies, high dimensionality, and the lack of an intuitive human-interpretable representation. We introduce MASCOTS, a method that leverages the Bag-of-Receptive-Fields representation alongside symbolic transformations inspired by Symbolic Aggregate Approximation. By operating in a symbolic feature space, it enhances interpretability while preserving fidelity to the original data and model. Unlike existing approaches that either depend on model structure or autoencoder-based sampling, MASCOTS directly generates meaningful and diverse counterfactual observations in a model-agnostic manner, operating on both univariate and multivariate data. We evaluate MASCOTS on univariate and multivariate benchmark datasets, demonstrating comparable validity, proximity, and plausibility to state-of-the-art methods, while significantly improving interpretability and sparsity. Its symbolic nature allows for explanations that can be expressed visually, in natural language, or through semantic representations, making counterfactual reasoning more accessible and actionable.

Problem

• Generates interpretable counterfactuals for time series models
• Handles temporal dependencies and high dimensionality challenges
• Works model-agnostically on univariate and multivariate data

Innovation

• Leverages Bag-of-Receptive-Fields representation
• Uses symbolic transformations for interpretability
• Model-agnostic counterfactual generation for time series