Smart Sensor Placement: A Correlation-Aware Attribution Framework (CAAF) for Real-world Data Modeling

2025/10/26 by Leung, Sze Chai, Zhou, Di, Bae, H. Jane
#Computational Engineering #FOS: Computer and information sciences #FOS: Electrical engineering #Finance #Machine Learning (cs.LG) #Systems and Control (eess.SY) #and Science (cs.CE) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2510.22517

Abstract

Optimal sensor placement (OSP) is critical for efficient, accurate monitoring, control, and inference in complex real-world systems. We propose a machine-learning-based feature attribution framework to identify OSP for the prediction of quantities of interest. Feature attribution quantifies input contributions to a model's output; however, it struggles with highly correlated input data often encountered in real-world applications. To address this, we propose a Correlation-Aware Attribution Framework (CAAF), which introduces a clustering step before performing feature attribution to reduce redundancy and enhance generalizability. We first illustrate the core principles of the proposed framework through a series of validation cases, then demonstrate its effectiveness in real-world dynamical systems, such as structural health monitoring, airfoil lift prediction, and wall-normal velocity estimation for turbulent channel flow. The results show that the CAAF outperforms alternative approaches that typically struggle due to the presence of nonlinear dynamics, chaotic behavior, and multi-scale interactions, and enables the effective application of feature attribution for identifying OSP in real-world environments.

Citations

• Deep reinforcement learning of airfoil pitch control in a highly disturbed environment using partial observations
• Environmental Sensor Placement with Convolutional Gaussian Neural Processes
• Algorithms to estimate Shapley value feature attributions
• Explainable AI: A Review of Machine Learning Interpretability Methods
• Optimal Experimental Design for Inverse Problems in the Presence of Observation Correlations
• Convolutional-network models to predict wall-bounded turbulence from wall quantities
• Interpreting Interpretations: Organizing Attribution Methods by Criteria
• A layered multiple importance sampling scheme for focused optimal Bayesian experimental design
• Explaining individual predictions when features are dependent: More accurate approximations to Shapley values
• Explaining individual predictions when features are dependent: More accurate approximations to Shapley values
• A Unified Approach to Interpreting Model Predictions
• Axiomatic Attribution for Deep Networks
• Temporal Convolutional Networks: A Unified Approach to Action Segmentation
• Sequential Bayesian optimal experimental design via approximate dynamic programming
• Lazier Than Lazy Greedy
• ImageNet Large Scale Visual Recognition Challenge
• ImageNet Large Scale Visual Recognition Challenge
• Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
• On the optimal sensor placement techniques for a bridge structure
• On the statistics of wind gusts
• Numerical solution of the Navier-Stokes equations

Related

• Optimal sensor placement using machine learning
• Measurement Optimization under Uncertainty using Deep Reinforcement Learning
• Low-Order Flow Reconstruction and Uncertainty Quantification in Disturbed Aerodynamics Using Sparse Pressure Measurements
• Human-Interpretable Uncertainty Explanations for Point Cloud Registration
• Correlation-Aware Feature Attribution Based Explainable AI