This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/107269
Description
Title
Automated theorem proving agent with memory
Author(s)
Zhang, Heling
Contributor(s)
Koyejo, Oluwasanmi
Issue Date
2020-05
Keyword(s)
automated theorem proving
deep learning
artificial intellegence
Abstract
Proof assistants are interactive software tools that performs automatic proof checking. Training
deep learning agents that are capable of interacting with proof assistants provide a novel way of
developing automated theorem proving (ATP) systems. We observe that existing methods in this
field, either supervised learning methods or RL, are all based on locality assumptions–the agent
generates proof tactics conditioning only on the information in the most recent time step, discarding
any information in previous time steps. However, the correctness of such assumptions has never
been verified. We propose a method that introduces global information in training deep-learning
based ATP models. Our contributions can be summarized as follows: (a) we construct a LSTM-like
memory mechanism across time steps for deep learning-based ATP agents, and (b) we build
a pipeline for end-to-end training by constructing a differentiable estimation of the proof assistant,
which can be viewed as a discrete function. We show the effectiveness of our method by building on
the state-of-the-art ATP model, ASTatics.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.