DeepAI AI Chat
Log In Sign Up

OoDAnalyzer: Interactive Analysis of Out-of-Distribution Samples

by   Changjian Chen, et al.

One major cause of performance degradation in predictive models is that the test samples are not well covered by the training data. Such not well-represented samples are called OoD samples. In this paper, we propose OoDAnalyzer, a visual analysis approach for interactively identifying OoD samples and explaining them in context. Our approach integrates an ensemble OoD detection method and a grid-based visualization. The detection method is improved from deep ensembles by combining more features with algorithms in the same family. To better analyze and understand the OoD samples in context, we have developed a novel kNN-based grid layout algorithm motivated by Hall's theorem. The algorithm approximates the optimal layout and has O(kN^2) time complexity, faster than the grid layout algorithm with overall best performance but O(N^3) time complexity. Quantitative evaluation and case studies were performed on several datasets to demonstrate the effectiveness and usefulness of OoDAnalyzer.


page 3

page 5

page 6

page 7

page 8

page 12

page 13

page 15


Fast and Work-Optimal Parallel Algorithms for Predicate Detection

Recently, the predicate detection problem was shown to be in the paralle...

ORC Layout: Adaptive GUI Layout with OR-Constraints

We propose a novel approach for constraint-based graphical user interfac...

Cluster-Aware Grid Layout

Grid visualizations are widely used in many applications to visually exp...

Does the Layout Really Matter? A Study on Visual Model Accuracy Estimation

In visual interactive labeling, users iteratively assign labels to data ...

Scalable Comparative Visualization of Ensembles of Call Graphs

Optimizing the performance of large-scale parallel codes is critical for...