Unified Explanations in Machine Learning Models: A Perturbation Approach

Simple item page
dc.contributor.authorDineen, Jacob
dc.contributor.authorKridel, Don
dc.contributor.authorDolk, Daniel
dc.contributor.authorCastillo, David
dc.date.accessioned2022-12-27T18:55:26Z
dc.date.available2022-12-27T18:55:26Z
dc.date.issued2023-01-03
dc.description.abstractA high-velocity paradigm shift towards Explainable Artificial Intelligence (XAI) has emerged in recent years. Highly complex Machine Learning (ML) models have flourished in many tasks of intelligence, and the questions have started to shift away from traditional metrics of validity towards something deeper: What is this model telling me about my data, and how is it arriving at these conclusions? Previous work has uncovered predictive models generating explanations contrasting domain experts, or excessively exploiting bias in data that renders a model useless in highly-regulated settings. These inconsistencies between XAI and modeling techniques can have the undesirable effect of casting doubt upon the efficacy of these explainability approaches. To address these problems, we propose a systematic, perturbation-based analysis against a popular, model-agnostic method in XAI, SHapley Additive exPlanations (Shap). We devise algorithms to generate relative feature importance in settings of dynamic inference amongst a suite of popular machine learning and deep learning methods, and metrics that allow us to quantify how well explanations generated under the static case hold. We propose a taxonomy for feature importance methodology, measure alignment, and observe quantifiable similarity amongst explanation models across several datasets.
dc.format.extent10
dc.identifier.doi10.24251/HICSS.2023.100
dc.identifier.isbn978-0-9981331-6-4
dc.identifier.othera8b055f1-df1b-47f8-b9b3-51a9a2e24157
dc.identifier.urihttps://hdl.handle.net/10125/102729
dc.language.isoeng
dc.relation.ispartofProceedings of the 56th Hawaii International Conference on System Sciences
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by-nc-nd/4.0/
dc.subjectBig Data and Analytics: Pathways to Maturity
dc.subjectartificial intelligence
dc.subjectdeep learning
dc.subjectexplainable artificial intelligence
dc.subjectmachine learning
dc.titleUnified Explanations in Machine Learning Models: A Perturbation Approach
dc.type.dcmitext
prism.startingpage795

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
0077.pdf
Size:
574.1 KB
Format:
Adobe Portable Document Format
Download

Collections

Big Data and Analytics: Pathways to Maturity