Output Reachable Set Estimation and Verification for Multilayer Neural Networks

In this brief, the output reachable estimation and safety verification problems for multilayer perceptron (MLP) neural networks are addressed. First, a conception called maximum sensitivity is introduced, and for a class of MLPs whose activation functions are monotonic functions, the maximum sensitivity can be computed via solving convex optimization problems. Then, using a simulation-based method, the output reachable set estimation problem for neural networks is formulated into a chain of optimization problems. Finally, an automated safety verification is developed based on the output reachable set estimation result. An application to the safety verification for a robotic arm model with two joints is presented to show the effectiveness of the proposed approaches.

[1]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[2]  Peter J. Gawthrop,et al.  Neural networks for control systems - A survey , 1992, Autom..

[3]  Steve W. Piche,et al.  The selection of weight accuracies for Madalines , 1995, IEEE Trans. Neural Networks.

[4]  Ah Chung Tsoi,et al.  Face recognition: a convolutional neural-network approach , 1997, IEEE Trans. Neural Networks.

[5]  Tao Zhang,et al.  Adaptive neural network control of nonlinear systems by state and output feedback , 1999, IEEE Trans. Syst. Man Cybern. Part B.

[6]  Daniel S. Yeung,et al.  Sensitivity analysis of multilayer perceptron to input and weight perturbations , 2001, IEEE Trans. Neural Networks.

[7]  Wing W. Y. Ng,et al.  Selection of weight quantisation accuracy for radial basis function neural network using stochastic sensitivity measure , 2003 .

[8]  Daniel S. Yeung,et al.  A Quantified Sensitivity Measure for Multilayer Perceptron to Input Perturbation , 2003, Neural Computation.

[9]  Daming Shi,et al.  Sensitivity analysis applied to the construction of radial basis function networks , 2005, Neural Networks.

[10]  Luca Pulina,et al.  An Abstraction-Refinement Approach to Verification of Artificial Neural Networks , 2010, CAV.

[11]  Luca Pulina,et al.  Challenging SMT solvers to verify neural networks , 2012, AI Commun..

[12]  Wang Xi-zhao,et al.  Architecture selection for networks trained with extreme learning machine using localized generalization error model , 2013 .

[13]  Xizhao Wang,et al.  Architecture selection for networks trained with extreme learning machine using localized generalization error model , 2013, Neurocomputing.

[14]  Yijing Wang,et al.  A non-ellipsoidal reachable set estimation for uncertain neural networks with time-varying delay , 2014, Commun. Nonlinear Sci. Numer. Simul..

[15]  Peng Shi,et al.  Exponential Stabilization for Sampled-Data Neural-Network-Based Control Systems , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[16]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[17]  Mahesh Viswanathan,et al.  C2E2: A Verification Tool for Stateflow Models , 2015, TACAS.

[18]  Xin Zhang,et al.  End to End Learning for Self-Driving Cars , 2016, ArXiv.

[19]  Weiming Xiang.  Necessary and Sufficient Condition for Stability of Switched Uncertain Linear Systems Under Dwell-Time Constraint , 2016, IEEE Transactions on Automatic Control.

[20]  Hieu Minh Trinh,et al.  Reachable sets bounding for generalized neural networks with interval time-varying delay and bounded disturbances , 2018, Neural Computing and Applications.

[21]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[22]  Mahesh Viswanathan,et al.  Automatic Reachability Analysis for Nonlinear Hybrid Models with C2E2 , 2016, CAV.

[23]  Jianbin Qiu,et al.  A Combined Adaptive Neural Network and Nonlinear Model Predictive Control for Multirate Networked Industrial Process Control , 2016, IEEE Transactions on Neural Networks and Learning Systems.

[24]  Weiming Xiang,et al.  On reachable set estimation for discrete-time switched linear systems under arbitrary switching , 2017, 2017 American Control Conference (ACC).

[25]  Weiming Xiang,et al.  Output Reachable Set Estimation for Switched Linear Systems and Its Application in Safety Verification , 2017, IEEE Transactions on Automatic Control.

[26]  James Lam,et al.  Stability analysis and L1-gain characterization for switched positive systems under dwell-time constraint , 2017, Autom..

[27]  Mykel J. Kochenderfer,et al.  Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks , 2017, CAV.

[28]  Weiming Xiang,et al.  Reachable Set Computation and Safety Verification for Neural Networks with ReLU Activations , 2017, ArXiv.

[29]  Min Wu,et al.  Safety Verification of Deep Neural Networks , 2016, CAV.

[30]  Stanley Bak,et al.  HyLAA: A Tool for Computing Simulation-Equivalent Reachability for Linear Systems , 2017, HSCC.

[31]  Stanley Bak,et al.  Rigorous Simulation-Based Analysis of Linear Hybrid Systems , 2017, TACAS.

[32]  Zheng-Guang Wu,et al.  Reachable Set Estimation for Markovian Jump Neural Networks With Time-Varying Delays , 2017, IEEE Transactions on Cybernetics.

[33]  Weiming Xiang,et al.  Parameter-memorized Lyapunov functions for discrete-time systems with time-varying parametric uncertainties , 2018, Autom..

[34]  Weiming Xiang,et al.  Robust Exponential Stability and Disturbance Attenuation for Discrete-Time Switched Systems Under Arbitrary Switching , 2018, IEEE Transactions on Automatic Control.