Saved in:
Bibliographic Details
Main Authors: Livanos, Michael, Davidson, Ian
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2403.18278
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916179850821632
author Livanos, Michael
Davidson, Ian
author_facet Livanos, Michael
Davidson, Ian
contents Deep learning is extensively used in many areas of data mining as a black-box method with impressive results. However, understanding the core mechanism of how deep learning makes predictions is a relatively understudied problem. Here we explore the notion of identifying a backbone of deep learning for a given group of instances. A group here can be instances of the same class or even misclassified instances of the same class. We view each instance for a given group as activating a subset of neurons and attempt to find a subgraph of neurons associated with a given concept/group. We formulate this problem as a set cover style problem and show it is intractable and presents a highly constrained integer linear programming (ILP) formulation. As an alternative, we explore a coverage-based heuristic approach related to pattern mining, and show it converges to a Pareto equilibrium point of the ILP formulation. Experimentally we explore these backbones to identify mistakes and improve performance, explanation, and visualization. We demonstrate application-based results using several challenging data sets, including Bird Audio Detection (BAD) Challenge and Labeled Faces in the Wild (LFW), as well as the classic MNIST data.
format Preprint
id arxiv_https___arxiv_org_abs_2403_18278
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Identification and Uses of Deep Learning Backbones via Pattern Mining
Livanos, Michael
Davidson, Ian
Artificial Intelligence
Deep learning is extensively used in many areas of data mining as a black-box method with impressive results. However, understanding the core mechanism of how deep learning makes predictions is a relatively understudied problem. Here we explore the notion of identifying a backbone of deep learning for a given group of instances. A group here can be instances of the same class or even misclassified instances of the same class. We view each instance for a given group as activating a subset of neurons and attempt to find a subgraph of neurons associated with a given concept/group. We formulate this problem as a set cover style problem and show it is intractable and presents a highly constrained integer linear programming (ILP) formulation. As an alternative, we explore a coverage-based heuristic approach related to pattern mining, and show it converges to a Pareto equilibrium point of the ILP formulation. Experimentally we explore these backbones to identify mistakes and improve performance, explanation, and visualization. We demonstrate application-based results using several challenging data sets, including Bird Audio Detection (BAD) Challenge and Labeled Faces in the Wild (LFW), as well as the classic MNIST data.
title Identification and Uses of Deep Learning Backbones via Pattern Mining
topic Artificial Intelligence
url https://arxiv.org/abs/2403.18278