| |
 Understanding Software Input Spaces Speaker (s):
Lars Grunske
Professor,
Department of Computer Science
Humboldt-Universität zu Berlin,
Germany
| Date: Time: Venue: | | 27 September 2024, Friday 11:00am – 12:00pm School of Computing &
Information Systems 2 (SCIS 2)
Level 4, Seminar Room 4-2
Singapore Management University
90 Stamford Road,
Singapore 178903 Please register by 25 September 2024. We look forward to seeing you at this research seminar. 
|
|
About the Talk A program's behavior, whether correct or not, is always influenced by specific program inputs. This leads to a general question: Which parts of the input and what specific (input) conditions trigger certain program behaviors, and how? This general question can be further refined into more specific ones that relate to our shared goal of automating software engineering tasks: 1) What is the precise set of inputs that trigger a certain bug, and how can it be characterized (eg. in terms of input constraints)? This would be relevant to better understand a faulty behavior during debugging. 2) What inputs lead to a certain line in the program being executed? This would help with regression testing and debugging/code understanding as well. 3) I have fixed a specific bug, but are there alternative execution paths that could lead to the same error location with a similar outcome? Identifying these paths would help detect incomplete bug fixes. 4-n) There are more of these question to be investigated.
To answer these questions, it is necessary to connect the characteristics of the input space to specific aspects of a program's execution. I assume that the program inputs are structured in some specification format, for example, using an input grammar. Based on this assumption, I will present current approaches and tools for grammar-based fuzzing and methods that automatically and incrementally infer the input conditions under which specific behaviors occur. Based on this, I will furthermore outline direction for further research in this area.
About the Speaker Lars Grunske is currently Professor at the Department of Computer Science from the Humboldt-Universität zu Berlin, Germany. He received his PhD degree in Computer Science from the University of Potsdam (Hasso-Plattner-Institute for Software Systems Engineering) in 2004. He was Boeing Postdoctoral Research Fellow at the University of Queensland from 2004-2007, a lecturer at the Swinburne University of Technology, Australia from 2008-2011, Junior Professor at the University of Kaiserslautern from 2011-2012, and Professor at the University of Stuttgart from 2012-2015. He has active research interests in automated software engineering and areas of modelling and verification of systems and software. His main focus is on automated software analysis, mainly probabilistic and timed model checking and model-based dependability evaluation of complex software intensive systems. His favored guitar is an Ibanez RG 1570 Prestige. |
