Principles for a Self-Explainable Model Through Information Theoretic Learning: Changkyu Choi (UiT) скачать в хорошем качестве

Principles for a Self-Explainable Model Through Information Theoretic Learning: Changkyu Choi (UiT)

Changkyu Choi, a Postdoctoral Research Fellow at the UiT Machine Learning Group and Visual Intelligence, gave a presentation titled "Formulating Explainability Principles for a Self-Explainable Model Through Information Theoretic Learning" on January 20th 2025 as part of the Visual Intelligence Online Seminar series. Changkyu Choi presented an approach in the field of document visual question answering (docVQA) that converts the conventional docVQA model into an inherently self-explainable system. Abstract: The recent development of self-explainable deep learning approaches has focused on integrating well-defined explainability principles into learning process, with the goal of achieving these principles through optimization. In this work, we propose DIB-X, a self-explainable deep learning approach for image data, which adheres to the principles of minimal, sufficient, and interactive explanations. The minimality and sufficiency principles are rooted from the trade-off relationship within the information bottleneck framework. Distinctly, DIB-X directly quantifies the minimality principle using the recently proposed matrix-based Rényi’s α-order entropy functional, circumventing the need for variational approximation and distributional assumption. The interactivity principle is realized by incorporating existing domain knowledge as prior explanations, fostering explanations that align with established domain understanding. As a preliminary result, we present an approach in the field of document visual question answering (docVQA) that converts the conventional docVQA model into inherently self-explainable system.