Universal Approximation with Certified Networks


Loading...

Date

2020-04

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric

Data

Abstract

Training neural networks to be certifiably robust is critical to ensure their safety against adversarial attacks. However, it is currently very difficult to train a neural network that is both accurate and certifiably robust. In this work we take a step towards addressing this challenge. We prove that for every continuous function f, there exists a network n such that: (i) n approximates f arbitrarily close, and (ii) simple interval bound propagation of a region B through n yields a result that is arbitrarily close to the optimal output of f on B. Our result can be seen as a Universal Approximation Theorem for interval-certified ReLU networks. To the best of our knowledge, this is the first work to prove the existence of accurate, interval-certified networks.

Publication status

published

Editor

Book title

Journal / series

Volume

Pages / Article No.

Publisher

International Conference on Learning Representations

Event

8th International Conference on Learning Representations (ICLR 2020) (virtual)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Adversarial; Adversarial attacks; Interval bound propagation; Relu networks; Robustness; Universal approximation

Organisational unit

03948 - Vechev, Martin / Vechev, Martin check_circle

Notes

Due to the Coronavirus (COVID-19) the conference was conducted virtually.

Funding

Related publications and datasets