FOSDEM proposal: Learning about Deep Learning: Applications for OpenJDK/Java Verification

[Shelley Lambert]

In this talk, we identify some well-known software verification problems,
using real-world examples from open-source projects and see how we might
apply some deep learning principles to address them.  In the various points
where we may test the Java runtime, we find candidates for deep learning.  What
is required?  We need a problem to solve, a model that describes it, and a
large amount of data to feed the neural network.


We will step through a simple example of where and how to apply deep
learning to more effectively test Java runtimes.  By covering the basics of
deep learning and the simple example, the intent of this presentation is to
spark curiosity and generate ideas on future applications of this machine
learning approach to problem-solving.


Recording: acceptable under CC-BY-2.0 license


Speaker Bio:

Shelley Lambert is the Test Lead for IBM Runtime Technologies team. She and
her team test open and freely available JDK implementations and have
delivered the test strategy, test code base, and test frameworks into the
Eclipse OMR, Eclipse OpenJ9 and AdoptOpenJDK projects. She is a committer
at OpenJ9 and AdoptOpenJDK and draws stories and lessons from her
experiences in the open projects where she is most active.


Blog URL: https://8thdaytesting.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.fosdem.org/pipermail/java-devroom/attachments/20181208/bba5e41e/attachment.html>