3 papers
cs.HC2025
Towards Difficulty-Aware Analysis of Deep Neural Networks
Linhao Meng, Stef van den Elzen, Anna Vilanova
Traditional instance-based model analysis focuses mainly on misclassified instances. However, this approach overlooks the varying difficulty associated with different instances. Id…
cs.LG2024
When Dimensionality Reduction Meets Graph (Drawing) Theory: Introducing a Common Framework, Challenges and Opportunities
Fernando Paulovich, Alessio Arleo, Stef van den Elzen
In the vast landscape of visualization research, Dimensionality Reduction (DR) and graph analysis are two popular subfields, often essential to most visual data analytics setups. D…
cs.HC2023
Class-constrained t-SNE: Combining Data Features and Class Probabilities
Linhao Meng, Stef van den Elzen, Nicola Pezzotti +1
Data features and class probabilities are two main perspectives when, e.g., evaluating model results and identifying problematic items. Class probabilities represent the likelihood…