[Pl-seminar] 3/24 Seminar: Swarat Chaudhuri, Learning to Program and Debug, Automatically

Daniel Patterson dbp at ccs.neu.edu
Sun Mar 12 12:14:37 EDT 2017


NUPRL Seminar presents

Swarat Chaudhuri
Rice University
Host: Amal Ahmed

12:00-1:30PM
Friday, March 24, 2017
Room 366 WVH (http://www.ccs.neu.edu/home/wand/directions.html)

Learning to Program and Debug, Automatically

Abstract:

Automating programming and debugging are long-standing goals in computer
science. In spite of significant progress in formal methods over the years,
we remain very far from achieving these goals. For example, a freshman CS
major will typically program circles around today's best program
synthesizers. Debugging and verification tools rely on formal
specifications, which are hard to provide in many important applications.

Two critical components of the gap between human and machine programmers
are that humans learn from experience, i.e., data, and can easily
generalize from incomplete problem definitions. In this talk, I will
present a general framework for formal methods, based on Bayesian
statistical learning, that aims to eliminate these differences. In our
framework, descriptions of programming tasks are seen to be "clues" towards
a hidden (probabilistic) specification that fully defines the task. Large
corpora of real-world programs are used to construct a statistical model
that correlates specifications with the form and function of their
implementations. The framework can be implemented in a variety of ways, but
in particular, through a neural architecture called Bayesian variational
encoder-decoders. Inferences made using the framework can be used to guide
traditional algorithms for program synthesis and bug-finding.

I will show that this data-driven approach can lead to giant leaps in the
scope and performance of automated program synthesis and debugging
algorithms. Specifically, I will give a demo of Bayou, a system for
Bayesian inductive synthesis of Java programs that goes significantly
beyond the state of the art in program synthesis. I will also describe
Salento, a debugging system based on our framework that can find subtle
violations of API contracts without any kind of specification.

Bio:

Swarat Chaudhuri is an Associate Professor of Computer Science at Rice
University. His research lies at the interface of programming systems and
artificial intelligence. Much of his recent work is on program synthesis,
the problem of automatically generating computer programs from high-level
specifications.
-------------- next part --------------
HTML attachment scrubbed and removed


More information about the pl-seminar mailing list