Finite-Time Model Inference From A Single Noisy Trajectory

This paper proposes a novel model inference procedure to identify system matrix from a single noisy trajectory over a finite-time interval. The proposed inference procedure comprises an observation data processor, a redundant data processor and an ordinary least-square estimator, wherein the data processors mitigate the influence of observation noise on inference error. We first systematically investigate the comparisons with naive least-square-regression based model inference and uncover that 1) the same observation data has identical influence on the feasibility of the proposed and the naive model inferences, 2) the naive model inference uses all of the redundant data, while the proposed model inference optimally uses the basis and the redundant data. We then study the sample complexity of the proposed model inference in the presence of observation noise, which leads to the dependence of the processed bias in the observed system trajectory on time and coordinates. Particularly, we derive the sample-complexity upper bound (on the number of observations sufficient to infer a model with prescribed levels of accuracy and confidence) and the sample-complexity lower bound (high-probability lower bound on model error). Finally, the proposed model inference is numerically validated and analyzed.

[1]  Dylan J. Foster,et al.  Learning nonlinear dynamical systems from a single trajectory , 2020, L4DC.

[2]  T. Lu,et al.  Inverses of 2 × 2 block matrices , 2002 .

[3]  Georgios B. Giannakis,et al.  Semi-Blind Inference of Topologies and Dynamical Processes Over Dynamic Graphs , 2018, IEEE Transactions on Signal Processing.

[4]  Tongxing Lu,et al.  Solution of the matrix equation AX−XB=C , 2005, Computing.

[5]  Csaba Szepesvári,et al.  Improved Algorithms for Linear Stochastic Bandits , 2011, NIPS.

[6]  Yongsoon Eun,et al.  A Multiplicative Coordinated Stealthy Attack and its Detection for Cyber Physical Systems , 2018, 2018 IEEE Conference on Control Technology and Applications (CCTA).

[7]  Xiaojing Zhang,et al.  Adaptive MPC for Iterative Tasks , 2018, 2018 IEEE Conference on Decision and Control (CDC).

[8]  M. Kanat Camlibel,et al.  Topology Reconstruction of Dynamical Networks via Constrained Lyapunov Equations , 2017, IEEE Transactions on Automatic Control.

[9]  Robert T. Pappalardo,et al.  Europa Clipper Mission Concept: Exploring Jupiter's Ocean Moon , 2014 .

[10]  Seibum Choi Antilock Brake System With a Continuous Wheel Slip Control to Maximize the Braking Performance and the Ride Quality , 2008, IEEE Transactions on Control Systems Technology.

[11]  M. Rudelson,et al.  Hanson-Wright inequality and sub-gaussian concentration , 2013 .

[12]  Divyakant Agrawal,et al.  Limiting the spread of misinformation in social networks , 2011, WWW.

[13]  Xiaojing Zhang,et al.  Adaptive MPC with Chance Constraints for FIR Systems , 2018, 2018 Annual American Control Conference (ACC).

[14]  Charles Baukal,et al.  Everything you need to know about nox , 2005 .

[15]  R. Adamczak,et al.  A note on the Hanson-Wright inequality for random vectors with dependencies , 2014, 1409.8457.

[16]  Steven Kay,et al.  Fundamentals Of Statistical Signal Processing , 2001 .

[17]  Minyue Fu,et al.  Exact Identification of Continuous-Time Systems from Sampled Data , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[18]  Richard H. Bartels,et al.  Algorithm 432 [C2]: Solution of the matrix equation AX + XB = C [F4] , 1972, Commun. ACM.

[19]  Patrik D'haeseleer,et al.  Genetic network inference: from co-expression clustering to reverse engineering , 2000, Bioinform..

[20]  Zhiwei Steven Wu,et al.  Random Quadratic Forms with Dependence: Applications to Restricted Isometry and Beyond , 2019, NeurIPS.

[21]  Alexandre Proutiere,et al.  Finite-time Identification of Stable Linear Systems Optimality of the Least-Squares Estimator , 2020, 2020 59th IEEE Conference on Decision and Control (CDC).

[22]  Roman Vershynin,et al.  High-Dimensional Probability , 2018 .

[23]  Paul-Marie Samson,et al.  Concentration of measure inequalities for Markov chains and $\Phi$-mixing processes , 2000 .

[24]  Naira Hovakimyan,et al.  Novel Stealthy Attack and Defense Strategies for Networked Control Systems , 2019, IEEE Transactions on Automatic Control.

[25]  Djalil Chafaï,et al.  Circular law for random matrices with exchangeable entries , 2014, Random Struct. Algorithms.

[26]  Kenneth E. Hibbard,et al.  Dragonfly: A rotorcraft lander concept for scientific exploration at titan , 2018 .

[27]  Samet Oymak,et al.  Non-asymptotic Identification of LTI Systems from a Single Trajectory , 2018, 2019 American Control Conference (ACC).

[28]  Alberto Bemporad,et al.  An MPC/hybrid system approach to traction control , 2006, IEEE Transactions on Control Systems Technology.

[29]  Mehran Mesbahi,et al.  Network Identification via Node Knockout , 2010, IEEE Transactions on Automatic Control.

[30]  Murti V. Salapaka,et al.  On the Problem of Reconstructing an Unknown Topology via Locality Properties of the Wiener Filter , 2010, IEEE Transactions on Automatic Control.

[31]  Ivan Markovsky Exact system identification with missing data , 2013, 52nd IEEE Conference on Decision and Control.

[32]  Jorge Cortés,et al.  Network Identification With Latent Nodes via Autoregressive Models , 2016, IEEE Transactions on Control of Network Systems.

[33]  Mohammad Al Hasan,et al.  A Survey of Link Prediction in Social Networks , 2011, Social Network Data Analytics.

[34]  Victor M. Preciado,et al.  Robust topology identification and control of LTI networks , 2014, 2014 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[35]  Duy Nguyen-Tuong,et al.  Local Gaussian Process Regression for Real Time Online Model Learning , 2008, NIPS.

[36]  Claire J. Tomlin,et al.  Sparse network identifiability via Compressed Sensing , 2016, Autom..

[37]  Manfred Morari,et al.  Learning and Control Using Gaussian Processes , 2018, 2018 ACM/IEEE 9th International Conference on Cyber-Physical Systems (ICCPS).

[38]  Jirí Matousek,et al.  On variants of the Johnson–Lindenstrauss lemma , 2008, Random Struct. Algorithms.

[39]  Munther A. Dahleh,et al.  Finite-Time System Identification for Partially Observed LTI Systems of Unknown Order , 2019, ArXiv.

[40]  Shahin Shahrampour,et al.  Topology Identification of Directed Dynamical Networks via Power Spectral Analysis , 2013, IEEE Transactions on Automatic Control.

[41]  Bernard Chazelle,et al.  Approximate nearest neighbors and the fast Johnson-Lindenstrauss transform , 2006, STOC '06.

[42]  Donatello Materassi,et al.  Topological identification in networks of dynamical systems , 2008, 2008 47th IEEE Conference on Decision and Control.

[43]  Alexander Rakhlin,et al.  Near optimal finite time identification of arbitrary linear dynamical systems , 2018, ICML.

[44]  Michael I. Jordan,et al.  Learning Without Mixing: Towards A Sharp Analysis of Linear System Identification , 2018, COLT.