Data-Induced Interactions of Sparse Sensors

Large-dimensional empirical data in science and engineering frequently has low-rank structure and can be represented as a combination of just a few eigenmodes. Because of this structure, we can use just a few spatially localized sensor measurements to reconstruct the full state of a complex system. The quality of this reconstruction, especially in the presence of sensor noise, depends significantly on the spatial configuration of the sensors. Multiple algorithms based on gappy interpolation and QR factorization have been proposed to optimize sensor placement. Here, instead of an algorithm that outputs a singular"optimal"sensor configuration, we take a thermodynamic view to compute the full landscape of sensor interactions induced by the training data. The landscape takes the form of the Ising model in statistical physics, and accounts for both the data variance captured at each sensor location and the crosstalk between sensors. Mapping out these data-induced sensor interactions allows combining them with external selection criteria and anticipating sensor replacement impacts.

[1]  Jaron Kent-Dobias.  Log-correlated color in Monet's paintings , 2022, 2209.01989.

[2]  J. Kutz,et al.  Data-driven sensor placement with shallow decoder networks , 2022, 2202.05330.

[3]  Sawyer B. Fuller,et al.  The Python Control Systems Library (python-control) , 2021, 2021 60th IEEE Conference on Decision and Control (CDC).

[4]  Michael C. Abbott,et al.  Information geometry for multiparameter models: new perspectives on the origin of simplicity , 2021, Reports on progress in physics. Physical Society.

[5]  Zhuoyuan Song,et al.  Dynamic Compressed Sensing of Unsteady Flows with a Mobile Robot , 2021, 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[6]  M. Brenner,et al.  Topological Design of Heterogeneous Self-Assembly , 2021, 2103.02010.

[7]  Steven L. Brunton,et al.  PySensors: A Python Package for Sparse Sensor Placement , 2021, J. Open Source Softw..

[8]  Samuel E. Otto,et al.  Inadequacy of Linear Methods for Minimal Sensor Placement and Feature Selection in Nonlinear Systems: A New Approach Using Secants , 2021, Journal of Nonlinear Science.

[9]  Thomas M. Smith,et al.  Improvements of the Daily Optimum Interpolation Sea Surface Temperature (DOISST) Version 2.1 , 2020, Journal of Climate.

[10]  Z. Drmač,et al.  Stability of Discrete Empirical Interpolation and Gappy Proper Orthogonal Decomposition with Randomized and Deterministic Sampling Points , 2020, SIAM J. Sci. Comput..

[11]  He Sun,et al.  Learning a Probabilistic Strategy for Computational Imaging Sensor Selection , 2020, 2020 IEEE International Conference on Computational Photography (ICCP).

[12]  Steven L. Brunton,et al.  Greedy Sensor Placement With Cost Constraints , 2018, IEEE Sensors Journal.

[13]  A. Klishin,et al.  Statistical physics of design , 2017, New Journal of Physics.

[14]  Madeleine Udell,et al.  Why Are Big Data Matrices Approximately Low Rank? , 2017, SIAM J. Math. Data Sci..

[15]  Steven L. Brunton,et al.  Data-Driven Sparse Sensor Placement for Reconstruction: Demonstrating the Benefits of Exploiting Known Patterns , 2017, IEEE Control Systems.

[16]  Bingni W. Brunton,et al.  Dynamic Mode Decomposition: Data-Driven Modeling of Complex Systems , 2016 .

[17]  George J. Pappas,et al.  Sensor placement for optimal Kalman filtering: Fundamental limits, submodularity, and algorithms , 2015, 2016 American Control Conference (ACC).

[18]  Arvind Murugan,et al.  Undesired usage and the robust self-assembly of heterogeneous structures , 2015, Nature Communications.

[19]  Scott Sheffield,et al.  Log-correlated Gaussian fields: an overview , 2014, 1407.5605.

[20]  Andrew Zisserman,et al.  Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps , 2013, ICLR.

[21]  Florent Krzakala,et al.  Statistical physics-based reconstruction in compressed sensing , 2011, ArXiv.

[22]  Andreas Krause,et al.  Submodularity and its applications in optimized information gathering , 2011, TIST.

[23]  Surya Ganguli,et al.  Statistical mechanics of compressed sensing. , 2010, Physical review letters.

[24]  Danny C. Sorensen,et al.  Discrete Empirical Interpolation for nonlinear model reduction , 2009, Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference.

[25]  W. Bialek,et al.  Statistical thermodynamics of natural images. , 2008, Physical review letters.

[26]  KrauseAndreas,et al.  Near-Optimal Sensor Placements in Gaussian Processes: Theory, Efficient Algorithms and Empirical Studies , 2008 .

[27]  T. Colonius,et al.  A fast immersed boundary method using a nullspace approach and multi-domain far-field boundary conditions , 2008 .

[28]  Tim Colonius,et al.  The immersed boundary method: A projection approach , 2007, J. Comput. Phys..

[29]  N. Nguyen,et al.  An ‘empirical interpolation’ method: application to efficient reduced-basis discretization of partial differential equations , 2004 .

[30]  Lawrence Sirovich,et al.  Karhunen–Loève procedure for gappy data , 1995 .

[31]  Andy Harter,et al.  Parameterisation of a stochastic model for human face identification , 1994, Proceedings of 1994 IEEE Workshop on Applications of Computer Vision.

[32]  Adrian S. Lewis,et al.  Image compression using the 2-D wavelet transform , 1992, IEEE Trans. Image Process..

[33]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[34]  M. L. Fisher,et al.  An analysis of approximations for maximizing submodular set functions—I , 1978, Math. Program..

[35]  H. Saruwatari,et al.  Region-Restricted Sensor Placement Based on Gaussian Process for Sound Field Estimation , 2022, IEEE Transactions on Signal Processing.

[36]  D. Donoho,et al.  The Optimal Hard Threshold for Singular Values is 4 / √ 3 , 2013 .

[37]  Jennifer Seberry,et al.  D-optimal designs , 2011 .