scholarly journals Embedding and classifying test execution traces using neural networks

IET Software ◽  
2021 ◽  
Author(s):  
Foivos Tsimpourlas ◽  
Gwenyth Rooijackers ◽  
Ajitha Rajan ◽  
Miltiadis Allamanis
Keyword(s):  
Neural Networks ◽  
Test Execution ◽  
Execution Traces

scholarly journals Evaluating distributed real-time and embedded system test correctness using system execution traces

2011 ◽  
Vol 1 (2) ◽  
Author(s):  
James Hill ◽  
Pooja Varshneya ◽  
Douglas Schmidt
Keyword(s):  
Embedded System ◽  
Real Time ◽  
Testing System ◽  
Software Components ◽  
System Quality ◽  
Test Quality ◽  
Trade Off ◽  
Test Execution ◽  
Execution Traces

AbstractEffective validation of distributed real-time and embedded (DRE) system quality-of-service (QoS) properties (e.g., event prioritization, latency, and throughput) requires testing system capabilities in representative execution environments. Unfortunately, evaluating the correctness of such tests is hard since it requires validating many states dispersed across many hardware and software components. To address this problem, this article presents a method called Test Execution (TE) Score for validating execution correctness of DRE system tests and empirically evaluates TE Score in the context of a representative DRE system. Results from this evaluation show that TE Score can determine the percentage correctness in test execution, and facilitate trade-off analysis of execution states, thereby increasing confidence in QoS assurance and improving test quality.

scholarly journals Verification of RNN-Based Neural Agent-Environment Systems

2019 ◽  
Vol 33 ◽  
pp. 6006-6013 ◽  
Author(s):  
Michael E. Akintunde ◽  
Andreea Kevorchian ◽  
Alessio Lomuscio ◽  
Edoardo Pirovano
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Recurrent Neural Network ◽  
Experimental Results ◽  
Feed Forward ◽  
Execution Traces ◽  
Verification Problem ◽  
Complete Procedure ◽  
Feed Forward Neural Networks

We introduce agent-environment systems where the agent is stateful and executing a ReLU recurrent neural network. We define and study their verification problem by providing equivalences of recurrent and feed-forward neural networks on bounded execution traces. We give a sound and complete procedure for their verification against properties specified in a simplified version of LTL on bounded executions. We present an implementation and discuss the experimental results obtained.

Sign in / Sign up

Export Citation Format

Share Document