[PRL] Fwd: [Programming] PL Seminar: Matt Might, "The War on Undecidability: Abstracting abstracting abstract machines"

[Benjamin Greenman]

---------- Forwarded message ----------
From: Stephen Chong <chong at seas.harvard.edu>
Date: Fri, Nov 6, 2015 at 7:06 PM
Subject: [Programming] PL Seminar: Matt Might, "The War on Undecidability:
Abstracting abstracting abstract machines"
To: "programming at eecs.harvard.edu" <programming at eecs.harvard.edu>


Hi all,
 Matt Might will be speaking this Wednesday, November 11, 4pm in MD 323.
Hope to see you there!

Cheers,
Steve.


Title: The War on Undecidability: Abstracting abstracting abstract machines
Speaker: Matt Might, University of Utah
Time: Wednesday November 11, 4pm
Place: MD 323

Predicting the behavior of programs has been “impossible” since Turing
first formulated the Halting Problem, the bedrock result of
theoretical computer science that gave rise to the notion of
undecidability.  Yet, at the same, predicting the behavior of programs
has become critically necessary for the security and performance of
modern software systems.  To evade undecidability in practice, our
species has learned to craft static analyses that do “reasonably well”
at predicting and bounding the behavior of programs of concern to
humans.  In this talk, I’ll include a tutorial on a broad universal
framework for static analysis of programs by abstract interpretation:
abstracting abstract machines (AAM).  I’ll then use this framework to
illustrate recent results in “The War on Undecidability.”

_______________________________________________
Programming mailing list
Programming at eecs.harvard.edu
https://lists.eecs.harvard.edu/mailman/listinfo/programming
-------------- next part --------------
HTML attachment scrubbed and removed