nn-dependability-kit: Engineering Neural Networks for Safety-Critical Systems

11/16/2018
by   Chih-Hong Cheng, et al.
0

nn-dependability-kit is an open-source toolbox to support safety engineering of neural networks. The key functionality of nn-dependability-kit includes (a) novel dependability metrics for indicating sufficient elimination of uncertainties in the product life cycle, (b) formal reasoning engine for ensuring that the generalization does not lead to undesired behaviors, and (c) runtime monitoring for reasoning whether a decision of a neural network in operation time is supported by prior similarities in the training data.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/05/2019

Testing and verification of neural-network-based safety-critical control software: A systematic literature review

Context: Neural Network (NN) algorithms have been successfully adopted i...
research
09/18/2018

Runtime Monitoring Neural Activation Patterns

For using neural networks in safety critical domains, it is important to...
research
07/19/2022

Using Neural Networks by Modelling Semi-Active Shock Absorber

A permanently increasing number of on-board automotive control systems r...
research
09/18/2018

Runtime Monitoring Neuron Activation Patterns

For using neural networks in safety critical domains, it is important to...
research
02/13/2023

EnergyShield: Provably-Safe Offloading of Neural Network Controllers for Energy Efficiency

To mitigate the high energy demand of Neural Network (NN) based Autonomo...
research
05/17/2018

DNN or k-NN: That is the Generalize vs. Memorize Question

This paper studies the relationship between the classification performed...
research
05/01/2023

Towards a Phenomenological Understanding of Neural Networks: Data

A theory of neural networks (NNs) built upon collective variables would ...

Please sign up or login with your details

Forgot password? Click here to reset