"Concolic Testing with Higher-Order Inputs"
21 views
Skip to first unread message
'John Clements' via users-redirect
May 18, 2020, 3:11:02 PM5/18/20
to Racket Users
Am I the first one to post about this? Looks like Robby’s giving a talk for the Chalmers online functional programming seminar series!
http://chalmersfp.org/
(and yes, this URL is non-ssl only, sigh)
Date: June 15th, 7am PDT / 10am EDT / 1400 UTC / 16:00 CEST
Title: “Concolic Testing with Higher-Order Inputs”
Abstract:
Concolic testing, dating back to the mid 2000s, has proven to be an effective, automatic testing technique. Roughly, a concolic tester monitors the program's behavior, collecting information that connects the branches that the program takes with the original inputs to the program. It then uses that information (with the help of an SMT solver) to try to construct new inputs to force the program to take different branches, in an attempt to uncover problems with the program.
Most of the existing work on concolic testing focuses on the situation where the inputs to the program being tested are flat values (e.g., numbers) instead of higher-order values (e.g., objects or functions). Higher-order inputs, however, present challenges for concolic testing because the interplay between the input and the program being tested is much more complex.
In this talk, I'll discuss the way we generalize concolic testing to higher-order inputs, explaining why it is challenging and how we answer the challenge. The talk will be example driven and I will try to bring across the intuition for how we conceptualize the generalization to higher-order inputs and the results we have so far.
I admit that I haven’t caught either of the first two talks in this series… apparently the Simon Peyton Jones one had some serious technical difficulties. Anyhow, those of us on the West Coast are probably getting more and more used to waking up early for global meetings, and this one certainly looks like a good one.
Can’t wait!
John
http://chalmersfp.org/
(and yes, this URL is non-ssl only, sigh)
Date: June 15th, 7am PDT / 10am EDT / 1400 UTC / 16:00 CEST
Title: “Concolic Testing with Higher-Order Inputs”
Abstract:
Concolic testing, dating back to the mid 2000s, has proven to be an effective, automatic testing technique. Roughly, a concolic tester monitors the program's behavior, collecting information that connects the branches that the program takes with the original inputs to the program. It then uses that information (with the help of an SMT solver) to try to construct new inputs to force the program to take different branches, in an attempt to uncover problems with the program.
Most of the existing work on concolic testing focuses on the situation where the inputs to the program being tested are flat values (e.g., numbers) instead of higher-order values (e.g., objects or functions). Higher-order inputs, however, present challenges for concolic testing because the interplay between the input and the program being tested is much more complex.
In this talk, I'll discuss the way we generalize concolic testing to higher-order inputs, explaining why it is challenging and how we answer the challenge. The talk will be example driven and I will try to bring across the intuition for how we conceptualize the generalization to higher-order inputs and the results we have so far.
I admit that I haven’t caught either of the first two talks in this series… apparently the Simon Peyton Jones one had some serious technical difficulties. Anyhow, those of us on the West Coast are probably getting more and more used to waking up early for global meetings, and this one certainly looks like a good one.
Can’t wait!
John
Reply all
Reply to author
Forward
0 new messages