Toward Reliable Neural Specifications

10/28/2022
by   Chuqin Geng, et al.
0

Having reliable specifications is an unavoidable challenge in achieving verifiable correctness, robustness, and interpretability of AI systems. Existing specifications for neural networks are in the paradigm of data as specification. That is, the local neighborhood centering around a reference input is considered to be correct (or robust). However, our empirical study shows that such a specification is extremely overfitted since usually no data points from the testing set lie in the certified region of the reference input, making them impractical for real-world applications. We propose a new family of specifications called neural representation as specification, which uses the intrinsic information of neural networks - neural activation patterns (NAP), rather than input data to specify the correctness and/or robustness of neural network predictions. We present a simple statistical approach to mining dominant neural activation patterns. We analyze NAPs from a statistical point of view and find that a single NAP can cover a large number of training and testing data points whereas ad hoc data-as-specification only covers the given reference data point. To show the effectiveness of discovered NAPs, we formally verify several important properties, such as various types of misclassifications will never happen for a given NAP, and there is no-ambiguity between different NAPs. We show that by using NAP, we can verify the prediction of the entire input space, while still recalling 84 argue that using NAPs is a more reliable and extensible specification for neural network verification.

READ FULL TEXT

page 2

page 7

page 14

page 15

page 16

page 17

page 18

research
06/21/2023

Verifying Global Neural Network Specifications using Hyperproperties

Current approaches to neural network verification focus on specification...
research
06/18/2020

Effective Formal Verification of Neural Networks using the Geometry of Linear Regions

Neural Networks (NNs) have increasingly apparent safety implications com...
research
06/24/2021

Online Verification of Deep Neural Networks under Domain or Weight Shift

Although neural networks are widely used, it remains challenging to form...
research
01/07/2019

Specification Patterns for Robotic Missions

Mobile and general-purpose robots increasingly support our everyday life...
research
02/23/2022

Learning Neural Networks under Input-Output Specifications

In this paper, we examine an important problem of learning neural networ...
research
11/09/2022

Enhanced Bayesian Neural Networks for Macroeconomics and Finance

We develop Bayesian neural networks (BNNs) that permit to model generic ...
research
02/24/2018

Time Series Learning using Monotonic Logical Properties

We propose a new paradigm for time-series learning where users implicitl...

Please sign up or login with your details

Forgot password? Click here to reset