Generative Code Modeling with Graphs

[kavon]

https://arxiv.org/abs/1805.08490

Generative Code Modeling with Graphs

Generative models for source code are an interesting structured prediction problem, requiring to reason about both hard syntactic and semantic constraints as well as about natural, likely programs. We present a novel model for this problem that uses a graph to represent the intermediate state of the generated output. The generative procedure interleaves grammar-driven expansion steps with graph augmentation and neural message passing steps. An experimental evaluation shows that our new model can generate semantically meaningful expressions, outperforming a range of strong baselines.

[brianhempel]

Haha, I see you've subscribed to the arXiv PL RSS?

I'm glad you added it here, it ended up in my "To Read" folder.

[kavon]

Yup! Caught my eye since it mentions graphs.

…

On Wed, 2018-05-23 at 09:56 -0700, Brian Hempel wrote: Haha, I see you've subscribed to the arXiv PL RSS? I'm glad you added it here, it ended up in my "To Read" folder. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.