Precise Multi-Neuron Abstractions for Neural Network Certification

by   Mark Niklas Müller, et al.

Formal verification of neural networks is critical for their safe adoption in real-world applications. However, designing a verifier which can handle realistic networks in a precise manner remains an open and difficult challenge. In this paper, we take a major step in addressing this challenge and present a new framework, called PRIMA, that computes precise convex approximations of arbitrary non-linear activations. PRIMA is based on novel approximation algorithms that compute the convex hull of polytopes, leveraging concepts from computational geometry. The algorithms have polynomial complexity, yield fewer constraints, and minimize precision loss. We evaluate the effectiveness of PRIMA on challenging neural networks with ReLU, Sigmoid, and Tanh activations. Our results show that PRIMA is significantly more precise than the state-of-the-art, verifying robustness for up to 16 than prior work on ReLU-, Sigmoid-, and Tanh-based networks, respectively.



page 1

page 2

page 3

page 4


A Tropical Approach to Neural Networks with Piecewise Linear Activations

We present a new, unifying approach following some recent developments o...

Reachable Set Computation and Safety Verification for Neural Networks with ReLU Activations

Neural networks have been widely used to solve complex real-world proble...

Exploiting Non-Linear Redundancy for Neural Model Compression

Deploying deep learning models, comprising of non-linear combination of ...

Validation of RELU nets with tropical polyhedra

This paper studies the problem of range analysis for feedforward neural ...

Neural Network Robustness Verification on GPUs

Certifying the robustness of neural networks against adversarial attacks...

Hierarchical Verification for Adversarial Robustness

We introduce a new framework for the exact point-wise ℓ_p robustness ver...

Scalable Verification of GNN-based Job Schedulers

Recently, Graph Neural Networks (GNNs) have been applied for scheduling ...

Code Repositories


ETH Robustness Analyzer for Deep Neural Networks

view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.