Marabou 2.0: A Versatile Formal Analyzer of Neural Networks

2024/01/25 by Wu, Haoze, Isac, Omri, Zeljić, Aleksandar +14 · 16 citations
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Logic in Computer Science (cs.LO) #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2401.14461

Abstract

This paper serves as a comprehensive system description of version 2.0 of the Marabou framework for formal analysis of neural networks. We discuss the tool's architectural design and highlight the major features and components introduced since its initial release.

Cited by

• Bridging Efficiency and Safety: Formal Verification of Neural Networks with Early Exits
• On Improving Deep Active Learning with Formal Verification
• Clip-and-Verify: Linear Constraint-Driven Domain Clipping for Accelerating Neural Network Verification
• Formal Analysis of the Sigmoid Function and Formal Proof of the Universal Approximation Theorem
• Proof Minimization in Neural Network Verification
• Verifying rich robustness properties for neural networks
• Probably Approximately Global Robustness Certification
• Floating-Point Neural Network Verification at the Software Level
• SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelism
• Additive Models Explained: A Computational Complexity Approach
• Probabilistic Robustness Analysis in High Dimensional Space: Application to Semantic Segmentation Network
• Scalable Synthesis and Verification of String Stable Neural Certificates for Interconnected Systems
• Formal Verification of Local Robustness of a Classification Algorithm for a Spatial Use Case
• Mini-Batch Robustness Verification of Deep Neural Networks
• Solution-aware vs global ReLU selection: partial MILP strikes back for DNN verification
• Of Good Demons and Bad Angels: Guaranteeing Safe Control under Finite Precision

Related

• SOCRATES: Towards a Unified Platform for Neural Network Analysis
• Neural Networks for Beginners. A fast implementation in Matlab, Torch, TensorFlow
• A Survey of Neural Networks and Formal Languages
• NEUROSPF: A tool for the Symbolic Analysis of Neural Networks
• Deep learning: Technical introduction