Distribution and Clusters Approximations as Abstract Domains in Probabilistic Abstract Interpretation to Neural Network Analysis
This work addresses neural network analysis for researchers in formal methods, but it appears incremental as it builds on existing frameworks without demonstrating broad improvements.
The paper tackles the problem of analyzing neural networks by introducing distribution and clusters approximations as abstract domains within the probabilistic abstract interpretation framework, showing how these methods work theoretically with abstract transformers and simple examples.
The probabilistic abstract interpretation framework of neural network analysis analyzes a neural network by analyzing its density distribution flow of all possible inputs. The grids approximation is one of abstract domains the framework uses which abstracts concrete space into grids. In this paper, we introduce two novel approximation methods: distribution approximation and clusters approximation. We show how these two methods work in theory with corresponding abstract transformers with help of illustrations of some simple examples.