# Dissecting Deep Neural Networks

arXiv: https://arxiv.org/abs/1910.03879 · HTML: https://mlsanity.com/paper/1910.03879/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Oct 2019.

## Built on by (most cited first, 7 of 7)

- [Traversing the Local Polytopes of ReLU Neural Networks: A Unified Approach for Network Verification](https://mlsanity.com/paper/2111.08922/) (2021)
- [FairProof : Confidential and Certifiable Fairness for Neural Networks](https://mlsanity.com/paper/2402.12572/) (2024)
- [Reachable Polyhedral Marching (RPM): An Exact Analysis Tool for Deep-Learned Control Systems](https://mlsanity.com/paper/2210.08339/) (2022)
- [Fast Jacobian-Vector Product for Deep Networks](https://mlsanity.com/paper/2104.00219/) (2021)
- [Instance-wise Linearization of Neural Network for Model Interpretation](https://mlsanity.com/paper/2310.16295/) (2023)
- [Parallel Algorithms for Exact Enumeration of Deep Neural Network Activation Regions](https://mlsanity.com/paper/2403.00860/) (2024)
- [Regional, Lattice and Logical Representations of Neural Networks](https://mlsanity.com/paper/2506.05834/) (2025)

## Abstract (arXiv)

In exchange for large quantities of data and processing power, deep neural networks have yielded models that provide state of the art predication capabilities in many fields. However, a lack of strong guarantees on their behaviour have raised concerns over their use in safety-critical applications. A first step to understanding these networks is to develop alternate representations that allow for further analysis. It has been shown that neural networks with piecewise affine activation functions are themselves piecewise affine, with their domains consisting of a vast number of linear regions. So far, the research on this topic has focused on counting the number of linear regions, rather than obtaining explicit piecewise affine representations. This work presents a novel algorithm that can compute the piecewise affine form of any fully connected neural network with rectified linear unit activations.