This tutorial explores Marabou, a tool for formally verifying deep neural networks (DNNs), and presents the theoretical foundations of DNN verification and an overview of Marabou’s core algorithms.
This work explores the application of solver-level warmstarting for neural network verification to exploit information from previous solutions, using a pipeline that is generalisable and potentially adaptable to state-of-the-art verifiers.
Annelot W. Bosman, Ming-Hao Liu, Marta Z. Kwiatkowska et al.· 0 citations
ZonoGpt, an abstract domain for verifying large transformers that maintains a space complexity independent of network depth, is introduced, an abstract domain for verifying large transformers that maintains a space complexity independent of network depth.
NNV3 is presented, the latest version of the Neural Network Verification tool, a MATLAB framework for formal verification of deep learning models and learning-enabled cyber-physical systems, demonstrating NNV's maturation into a comprehensive, robust, and accessible verification tool for a diverse range of AI systems.
Anne M. Tumlin, Samuel Sasaki, Ben Wooding et al.· 0 citations
HSRM is introduced, a lightweight hidden-state reward model that verifies candidate solutions by directly reading the generator's internal representations rather than re-processing its text, providing an efficient alternative to text-only verification by reusing representations already computed during generation.
Constrained Lagrangian Abstract Domain (CLAD), a new abstract domain that computes a sound over-approximation of neural networks over input regions defined by a combination of convex constraints, is introduced.