FORCE: Feature-Oriented Representation with Clustering and Explanation

2025-04-07

AI Summary

This paper addresses the insufficient modeling of latent variables in representation learning. We propose an interpretable deep learning framework that explicitly captures the influence of unobserved factors on feature importance via two-stage SHAP value integration. First, SHAP values serve as supervision signals to guide K-means clustering, yielding semantically interpretable latent features. Second, SHAP values are embedded into a learnable attention mechanism to enable dynamic modulation of feature weights by latent structure. To our knowledge, this is the first work to directly transform model-agnostic explanation scores into end-to-end differentiable architectural components, jointly achieving interpretability and discriminative power. Extensive experiments on three real-world healthcare datasets demonstrate efficacy: e.g., F1-score for heart disease prediction reaches 0.80—improving over baselines by 0.08—validating that synergistic integration of latent features and attention significantly enhances performance.

Technology Category

• Machine Learning: Transparent, Interpretable, Explainable ML
• Computer Vision: Interpretability, Explainability, and Transparency
• Humans and AI: Explainable AI (XAI) for Human Understanding

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Learning about underlying patterns in data using latent unobserved structures to improve the accuracy of predictive models has become an active avenue of deep learning research. Most approaches cluster the original features to capture certain latent structures. However, the information gained in the process can often be implicitly derived by sufficiently complex models. Thus, such approaches often provide minimal benefits. We propose a SHAP (Shapley Additive exPlanations) based supervised deep learning framework FORCE which relies on two-stage usage of SHAP values in the neural network architecture, (i) an additional latent feature to guide model training, based on clustering SHAP values, and (ii) initiating an attention mechanism within the architecture using latent information. This approach gives a neural network an indication about the effect of unobserved values that modify feature importance for an observation. The proposed framework is evaluated on three real life datasets. Our results demonstrate that FORCE led to dramatic improvements in overall performance as compared to networks that did not incorporate the latent feature and attention framework (e.g., F1 score for presence of heart disease 0.80 vs 0.72). Using cluster assignments and attention based on SHAP values guides deep learning, enhancing latent pattern learning and overall discriminative capability.

Problem

• Improves predictive models using latent feature clustering
• Enhances pattern learning via SHAP-based attention mechanisms
• Boosts discriminative capability in deep learning frameworks

Innovation

• SHAP-based supervised deep learning framework
• Clustering SHAP values for latent feature
• Attention mechanism with latent information