Skip to Main Content
 

Global Search Box

 
 
 

ETD Abstract Container

Abstract Header

Multi-Semantic-Stage Neural Networks for Robust and Interpretable Deep Learning

Menart, Christopher J.
http://rave.ohiolink.edu/etdc/view?acc_num=wright1734349340890461

Abstract Details

2024, Doctor of Philosophy (PhD), Wright State University, Computer Science and Engineering PhD.
Deep neural networks have great representational power. However, most deep neural nets today optimize directly for performance on a single task defined only by labeled training data. This excludes potential sources of knowledge and ways of learning which could improve their performance, and address challenges, such as explainability, which are pressing to the field. We propose a framework for neural network architecture which generalizes it to a graph of many semantically-meaningful variables. We call it the Multi-Semantic-Stage Neural Network (MSSNN). An MSSNN models its domain as a web of conditional probabilities, i.e. a collection of inter-related tasks which can learn from each other. This way of conceptualizing a deep learning model manifests in several immediately useful capabilities. Such a model can easily exploit more and different kinds of labels in its training data. It can supplement its training data by incorporating expert knowledge, both in the structure of its graph and in the form of hard-coded relationships between variables. Graph structures such as multiple incident edges on a single variable automatically lead to semi-supervised learning ability. And finally, the MSSNN’s ability to sample from learned joint distributions allows us to construct novel explanations, directly tied to actual causes of the model’s behavior. We perform an initial demonstration of these capabilities by constructing and evaluating Multi-Semantic-Stage Neural Networks of several sizes for a collection of computer vision tasks.
Michael Raymer, Ph.D. (Advisor)
Thomas Wischgoll, Ph.D. (Committee Member)
Cogan Shimizu, Ph.D. (Committee Member)
Olga Mendoza-Schrock, Ph.D. (Committee Member)
Vince Velten, Ph.D. (Committee Member)
Mateen Rizki, Ph.D. (Committee Member)
191 p.
Artificial Intelligence; Computer Science
Deep Learning; Belief Networks; Knowledge Engineering; Semi-Supervised Learning; Explainable AI

Recommended Citations

Citations

  • Menart, C. J. (2024). Multi-Semantic-Stage Neural Networks for Robust and Interpretable Deep Learning [Doctoral dissertation, Wright State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=wright1734349340890461

    APA Style (7th edition)

  • Menart, Christopher. Multi-Semantic-Stage Neural Networks for Robust and Interpretable Deep Learning. 2024. Wright State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=wright1734349340890461.

    MLA Style (8th edition)

  • Menart, Christopher. "Multi-Semantic-Stage Neural Networks for Robust and Interpretable Deep Learning." Doctoral dissertation, Wright State University, 2024. http://rave.ohiolink.edu/etdc/view?acc_num=wright1734349340890461

    Chicago Manual of Style (17th edition)

wright1734349340890461
52
© 2024, some rights reserved.Creative Commons License Multi-Semantic-Stage Neural Networks for Robust and Interpretable Deep Learning by Christopher J. Menart is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by Wright State University and OhioLINK.