Interactive Dense Pixel Visualizations for Time Series and Model Attribution Explanations

Simple item page
dc.contributor.authorSchlegel, Udoen_US
dc.contributor.authorKeim, Danielen_US
dc.contributor.editorArchambault, Danielen_US
dc.contributor.editorNabney, Ianen_US
dc.contributor.editorPeltonen, Jaakkoen_US
dc.date.accessioned2023-06-10T05:50:51Z
dc.date.available2023-06-10T05:50:51Z
dc.date.issued2023
dc.description.abstractThe field of Explainable Artificial Intelligence (XAI) for Deep Neural Network models develops significantly, offering numerous techniques to extract explanations from models. However, evaluating explanations is often not trivial, and differences in applied metrics can be subtle, especially with non-intelligible data. Thus, there is a need for visualizations tailored to explore explanations for domains with such data, e.g., time series. We propose DAVOTS, an interactive visual analytics approach to explore raw time series data, activations of neural networks, and attributions in a dense-pixel visualization to gain insights into the data, models' decisions, and explanations. To further support users in exploring large datasets, we apply clustering approaches to the visualized data domains to highlight groups and present ordering strategies for individual and combined data exploration to facilitate finding patterns. We visualize a CNN trained on the FordA dataset to demonstrate the approach.en_US
dc.description.seriesinformationMachine Learning Methods in Visualisation for Big Data
dc.identifier.doi10.2312/mlvis.20231113
dc.identifier.isbn978-3-03868-224-0
dc.identifier.pages1-5
dc.identifier.pages5 pages
dc.identifier.urihttps://doi.org/10.2312/mlvis.20231113
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/mlvis20231113
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Human-centered computing -> Visual analytics;Computing methodologies -> Neural networks
dc.subjectHuman centered computing
dc.subjectVisual analytics
dc.subjectComputing methodologies
dc.subjectNeural networks
dc.titleInteractive Dense Pixel Visualizations for Time Series and Model Attribution Explanationsen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
001-005.pdf
Size:
1.55 MB
Format:
Adobe Portable Document Format
Download
Collections
Machine Learning Methods in Visualisation for Big Data 2023