Discovering Classifier Errors with AnchorViz in Semantic Data Exploration
AnchorViz facilitates the discovery of classifier errors through interactive semantic data exploration, reducing feature blindness and improving robustness for critical systems. Learn how AnchorViz aligns with user strategies, alleviates blindness errors in machine learning models, and enhances human-computer interaction. Discover the relevance of AnchorViz in enhancing error detection, visualization for model interpretability, and facilitating active learning in team projects.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
AnchorViz: Facilitating Classifier Error Discovery through Interactive Semantic Data Exploration Discussant: Austin McCormick
Relevance of AnchorVizToday ML models increase in quantity and complexity Reducing Feature Blindness Human-Computer Interaction for Discovery and Explainability Simplifies Manual Inspection Improves Robustness for Critical Systems
How does AnchorVizalign with Simon? Satisficing over Optimizing Users use their own rules and strategies for exploring semantically similar data when creating anchors Complexity Caused by the Environment Feature blindness - a model s inability to predict data patterns due to lacking features or training data for those patterns Blindness errors occur when a model doesn t account for the full complexity of the environment and lacks features or training data AnchorViz alleviates blindness errors in ML models Decomposition Anchors break down the complex data environment into interpretable components
How does AnchorVizalign with Suchman? Human-Machine Collaboration AnchorViz is centered on human-computer interaction AnchorViz lets users interact with data by creating anchors Provides transparency and explainability for the refined model Situated Action Decisions shaped by dynamic context User action is dependent on the data visualization environment AnchorViz exploration isn t a pre-planned rigid script Anchor placement is dependent on the context of the user
Relevance to Team Project Enhance Error Detection Visualization used for model interpretability Facilitation of Human-Computer Interaction Reduction of Cognitive Load Facilitation of Active Learning
