Reference : Many-Objective Reinforcement Learning for Online Testing of DNN-Enabled Systems

	Document type :	Scientific congresses, symposiums and conference proceedings : Paper published in a book
	Discipline(s) :	Engineering, computing & technology : Computer science
	Focus Areas :	Security, Reliability and Trust
	To cite this reference:	http://hdl.handle.net/10993/55817

Title :	Many-Objective Reinforcement Learning for Online Testing of DNN-Enabled Systems
Language :	English
Author, co-author :	Ul Haq, Fitash [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SVV >]
	Shin, Donghwan [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SVV >]
	Briand, Lionel [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > SVV >]
Publication date :	May-2023
Main document title :	45th International Conference on Software Engineering (ICSE ’23)
Publisher :	ACM
Peer reviewed :	Yes
On invitation :	No
Audience :	International
City :	New York, NY
Country :	USA
Event name :	45th International Conference on Software Engineering (ICSE ’23)
Event date :	from 14-05-2023 to 20-05-2023
Keywords :	[en] DNN Testing ; Reinforcement learning ; Many objective search
Abstract :	[en] Deep Neural Networks (DNNs) have been widely used to perform real-world tasks in cyber-physical systems such as Autonomous Driving Systems (ADS). Ensuring the correct behavior of such DNN-Enabled Systems (DES) is a crucial topic. Online testing is one of the promising modes for testing such systems with their application environments (simulated or real) in a closed loop, taking into account the continuous interaction between the systems and their environments. However, the environmental variables (e.g., lighting conditions) that might change during the systems' operation in the real world, causing the DES to violate requirements (safety, functional), are often kept constant during the execution of an online test scenario due to the two major challenges: (1) the space of all possible scenarios to explore would become even larger if they changed and (2) there are typically many requirements to test simultaneously. In this paper, we present MORLOT (Many-Objective Reinforcement Learning for Online Testing), a novel online testing approach to address these challenges by combining Reinforcement Learning (RL) and many-objective search. MORLOT leverages RL to incrementally generate sequences of environmental changes while relying on many-objective search to determine the changes so that they are more likely to achieve any of the uncovered objectives. We empirically evaluate MORLOT using CARLA, a high-fidelity simulator widely used for autonomous driving research, integrated with Transfuser, a DNN-enabled ADS for end-to-end driving. The evaluation results show that MORLOT is significantly more effective and efficient than alternatives with a large effect size. In other words, MORLOT is a good option to test DES with dynamically changing environments while accounting for multiple safety requirements.
Research centres :	Interdisciplinary Centre for Security, Reliability and Trust (SnT) > SVV - Software Verification and Validation
Target :	Researchers ; Professionals ; Students ; General public
Permalink :	http://hdl.handle.net/10993/55817
DOI :	10.1109/ICSE48619.2023.00155

	File(s) associated to this reference
	Fulltext file(s): File Commentary Version Size Access Open access ICSE48619.2023.00155.pdf Publisher postprint 691.05 kB View/Open

All documents in ORBi^lu are protected by a user license.

 

University of Luxembourg Library |  Feedback |  Legal notices Site Map