Sign up to save your podcasts
Or
Share Program Development by Proof Transformation
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
Program Development by Proof Transformation
In the last 20 years the formal approach to the development
of software turned out to be a crucial technique for the generation
of correct programs.
This idea has its theoretical base into the several semi-automatic
methods to transform a formal specification that describe the behavior
of a program into an effective executable piece of code.
One of this is the so-called "program extraction from proof". The idea
is that from an constructive proof of a formula "for each x there exists
y such that P(x,y)" we can automatically extract a program "t" such that
the property P(x,t(x)) hold. In our days such proofs are normally written
by ad-hoc tools (some of them are: COQ, ISABLLE, MINLOG, PX, AGDA, etc...)
called "proof assistants".
Even if today this technique is pretty well established, the "manipulation"
of proofs in order to develop performing programs did not received big attention.
In this thesis we will develop several automatic and semi-automatic methods in
order to extract efficient code from constructive proofs. Our field of application
will be computational biology, a research field in which the development of
efficient programs is crucial. So our main goal will be to show how the manipulation
of formal proofs, essentially studied by proof theorist, has a big effect also in
practical program generation.
View all episodes
By Ludwig-Maximilians-Universität München
5
11 ratings
In the last 20 years the formal approach to the development
of software turned out to be a crucial technique for the generation
of correct programs.
This idea has its theoretical base into the several semi-automatic
methods to transform a formal specification that describe the behavior
of a program into an effective executable piece of code.
One of this is the so-called "program extraction from proof". The idea
is that from an constructive proof of a formula "for each x there exists
y such that P(x,y)" we can automatically extract a program "t" such that
the property P(x,t(x)) hold. In our days such proofs are normally written
by ad-hoc tools (some of them are: COQ, ISABLLE, MINLOG, PX, AGDA, etc...)
called "proof assistants".
Even if today this technique is pretty well established, the "manipulation"
of proofs in order to develop performing programs did not received big attention.
In this thesis we will develop several automatic and semi-automatic methods in
order to extract efficient code from constructive proofs. Our field of application
will be computational biology, a research field in which the development of
efficient programs is crucial. So our main goal will be to show how the manipulation
of formal proofs, essentially studied by proof theorist, has a big effect also in
practical program generation.
More shows like Fakultät für Mathematik, Informatik und Statistik - Digitale Hochschulschriften der LMU - Teil 01/02
View all
Theoretical Physics Schools (ASC)
2 Listeners
Katholisch-Theologische Fakultät - Digitale Hochschulschriften der LMU
0 Listeners
MCMP – Mathematical Philosophy (Archive 2011/12)
6 Listeners
Hegel lectures by Robert Brandom, LMU Munich
6 Listeners
John Lennox - Hat die Wissenschaft Gott begraben?
3 Listeners
MCMP – Philosophy of Science
2 Listeners
MCMP – Philosophy of Mathematics
2 Listeners
Epistemology and Philosophy of Science: Prof. Dr. Stephan Hartmann – HD
1 Listeners
MCMP – Philosophy of Physics
4 Listeners
Center for Advanced Studies (CAS) Research Focus Evolutionary Biology (LMU) - HD
0 Listeners