Tang, Haoteng
(2023)
Interpretable Graph Representation Learning: New Theories and Applications.
Doctoral Dissertation, University of Pittsburgh.
(Unpublished)
Abstract
Recent decades have witnessed the prosperity of deep learning which has revolutionized a broad variety of fields, ranging from computer vision recognition to natural language understanding. Although deep learning has achieved great success on Euclidean data (e.g., images, language sequences), the studies and explorations of the deep learning methods on graph-structured data are far from enough. The graph-structured data, presenting the relations among different items, are ubiquitous in the real world, such as transportation networks, social networks, and biological networks. However, it is challenging for regular deep learning methods to capture the hierarchical structures rooted in the graph-structured data. Another limitation of the previous graph learning models is that most of them mainly focus on unsigned graphs (i.e., graphs that only include positive and negative edges) learning. Beyond these, most of the current graph learning models are not interpretable. To address these issues, new interpretable deep graph learning models are proposed for both signed and unsigned graphs to capture the hierarchical structures in graphs and yield whole graph representations for graph-level tasks (i.e., graph classifications, and regressions). Several graph-related applications are also presented to show the practical merits of graph-structured data to the AI community.
Share
Citation/Export: |
|
Social Networking: |
|
Details
Item Type: |
University of Pittsburgh ETD
|
Status: |
Unpublished |
Creators/Authors: |
|
ETD Committee: |
|
Date: |
14 September 2023 |
Date Type: |
Publication |
Defense Date: |
31 March 2023 |
Approval Date: |
14 September 2023 |
Submission Date: |
2 June 2023 |
Access Restriction: |
No restriction; Release the ETD for access worldwide immediately. |
Number of Pages: |
143 |
Institution: |
University of Pittsburgh |
Schools and Programs: |
Swanson School of Engineering > Electrical and Computer Engineering |
Degree: |
PhD - Doctor of Philosophy |
Thesis Type: |
Doctoral Dissertation |
Refereed: |
Yes |
Uncontrolled Keywords: |
brain networks, graph reasoning, semantic segmentation, graph representation learning, hierarchical structure, interpretable model, signed and unsigned graphs |
Date Deposited: |
14 Sep 2023 13:37 |
Last Modified: |
14 Sep 2023 13:37 |
URI: |
http://d-scholarship.pitt.edu/id/eprint/44932 |
Metrics
Monthly Views for the past 3 years
Plum Analytics
Actions (login required)
![View Item View Item](/style/images/action_view.png) |
View Item |