Verification for Machine Learning, Autonomy, and Neural Networks Survey

Abstract

This survey presents an overview of verification techniques for autonomous systems, with a focus on safety-critical autonomous cyber-physical systems (CPS) and subcomponents thereof. Autonomy in CPS is enabling by recent advances in artificial intelligence (AI) and machine learning (ML) through approaches such as deep neural networks (DNNs), embedded in so-called learning enabled components (LECs) that accomplish tasks from classification to control. Recently, the formal methods and formal verification community has developed methods to characterize behaviors in these LECs with eventual goals of formally verifying specifications for LECs, and this article presents a survey of many of these recent approaches.

References in corpus (18)

• Intriguing properties of neural networks
• End to End Learning for Self-Driving Cars
• DeepXplore: Automated Whitebox Testing of Deep Learning Systems
• An approach to reachability analysis for feed-forward ReLU neural networks
• Approximate Optimal Trajectory Tracking for Continuous Time Nonlinear Systems
• DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Cars
• Testing Deep Neural Networks
• A Dual Approach to Scalable Verification of Deep Networks
• Towards Fast Computation of Certified Robustness for ReLU Networks
• Efficient model-based reinforcement learning for approximate online optimal
• Reinforcement Learning: A Survey
• Output Range Analysis for Deep Neural Networks
• Neural-Guided Deductive Search for Real-Time Program Synthesis from Examples
• Reachability-Based Safety and Goal Satisfaction of Unmanned Aerial Platoons on Air Highways
• Deep Reinforcement Learning for Autonomous Driving
• Safe Platooning of Unmanned Aerial Vehicles via Reachability
• Reachable Set Estimation and Safety Verification for Piecewise Linear Systems with Neural Network Controllers
• Reachable Set Estimation and Verification for Neural Network Models of Nonlinear Dynamic Systems

Cited by in corpus (14)

• Forecasting remaining useful life: Interpretable deep learning approach via variational Bayesian inferences
• A Roadmap Towards Resilient Internet of Things for Cyber-Physical Systems
• Machine Learning Testing: Survey, Landscapes and Horizons
• Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges
• Adversarial Robustness of Deep Neural Networks: A Survey from a Formal Verification Perspective
• Algorithms for Verifying Deep Neural Networks
• Energy-Efficient Control Adaptation with Safety Guarantees for Learning-Enabled Cyber-Physical Systems
• A Survey of Recent Scalability Improvements for Semidefinite Programming with Applications in Machine Learning, Control, and Robotics
• Verification of Sigmoidal Artificial Neural Networks using iSAT
• Formal controller synthesis for hybrid systems using genetic programming
• ReLUSyn: Synthesizing Stealthy Attacks for Deep Neural Network Based Cyber-Physical Systems
• The Final Frontier: Deep Learning in Space
• Reachability Analysis of Neural Networks with Uncertain Parameters
• Formal Verification of Neural Network Controlled Autonomous Systems