University of Notre Dame
Browse
CARMICHAELZJ42024D.pdf (13.36 MB)

Explainable AI for High-Stakes Decision-Making

Download (13.36 MB)
dataset
posted on 2024-05-01, 18:21 authored by Zachariah Carmichael
Canonical AI algorithms are black boxes. When it comes to high-stakes applications, this is highly undesirable—we need algorithms that we can understand and, in turn, trust. Doctors will not, and should not, trust a machine that cannot reason about its decisions. The push for this explainability has led to the emergence of the explainable AI (XAI) sub-field. While XAI has made progress over the years, there still exists a wealth of issues. In this dissertation, we demonstrate a path toward the glass box ideal: highly performant algorithms that are fully human-comprehensible. We first demonstrate that post hoc explainers, while incredibly popular, should not be trusted. In certain cases in which they must be used, we provide a solution that helps the explainers become more reliable. We characterize the open problems of building an AI system feasible for high-stakes applications through the design of a real-world emergency drone response system. This motivates us to challenge the fundamentals of deep learning architectural design. We bridge XAI and automated machine learning (AutoML) to discover intrinsically debuggable neural networks. Next, we propose novel, intrinsically interpretable approaches to computer vision based on prototypical networks, a type of concept-based neural network. Critically, we address the human-machine semantic similarity gap associated with learned prototypes. First, we enable the automatic learning of prototypical parts with weak supervision using receptive-field constrained networks. Second, we enrich the interpretability of learned concepts by learning prototypical distributions in the invertible latent space of a normalizing flow. We demonstrate a measurable improvement in the comprehensibility of decisions made in both predictive and generative computer vision tasks. The contributions of this dissertation open up the AI black box by substituting correlative explainers with faithful, actionable, intuitive, and debuggable algorithms.

History

Date Created

2024-04-07

Date Modified

2024-05-01

Defense Date

2024-03-27

CIP Code

  • 14.0901

Research Director(s)

Walter Scheirer

Committee Members

Kevin Bowyer Adam Czajka Michael Jones

Degree

  • Doctor of Philosophy

Degree Level

  • Doctoral Dissertation

Language

  • English

Library Record

006583036

OCLC Number

1432241324

Publisher

University of Notre Dame

Program Name

  • Computer Science and Engineering

Usage metrics

    Dissertations

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC