Award Date

12-1-2021

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Science

First Committee Member

Kazem Taghva

Second Committee Member

Mingon Kang

Third Committee Member

Fatma Nasoz

Fourth Committee Member

Henry Selvaraj

Number of Pages

76

Abstract

Language comprehension or more formally, natural language understanding is one of the major undertakings in Artificial Intelligence. In this work, we explore a few of the problems in language understanding using fixed deep learning models. Specifically, first, we look into question generation. Asking questions relates to the cognitive ability of language comprehension and context understanding. For that reason, making progress in question generation is significant. We introduce a novel task called “question generation with masked target answer” and propose various models and present the baseline result for the task. Next, we extend on the question generation task and develop a large-scale dataset for our task and for question generation in general. Next, we explore the problem of paraphrase identification, in which the task is to decide whether a pair of sentences is a paraphrase of each other. We present various machine learning models and discuss their performance. Moving on from the fixed architecture of deep learning models, we then explore the area of neuroevolution where the models constantly change based on some evolutionary operators and learn until an optimal architecture is found. This direction promises to create a more general form of intelligence. In particular, we formulate a recombination algorithm called Highest Varying k-Features Recombination(HVk-FR) and use it on top of various mutation operators to evolve the models. We show how our proposed algorithm can actually go in the direction of optimal network structure starting from a basic one-layer deep network.

Keywords

Deep learning; General AI; Natural language processing; Neuroevolution

Disciplines

Computer Sciences

File Format

pdf

File Size

1479 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Rights

IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/


Share

COinS