Interactive Dimensionality Reduction for Comparative Analysis

Fujiwara, Takanori; Wei, Xinhai; Zhao, Jian; Ma, Kwan‐Liu

doi:10.1109/tvcg.2021.3114807

Cited by 27 publications

(7 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…How to plug visual analysis tool into experts' development environment. ULCA [21] and PipelineProfiler [57] targeting simple models and datasets are integrated with Jupyter Notebook, and a few other works focusing on deep models, such as ex-plAIner [71], are TensorBoard [1] plug-ins. The target users of our work are experts developing QA models, who generally use Python and are familiar with TensorBoard.…”

Section: Discussionmentioning

confidence: 99%

“…Embedding Analysis. The combination of Dimension Reduction (DR) algorithms and visualization is a common method to analyze embeddings [21] and offering interpretability [22]. In the NLP context, researchers usually use PCA [27], t-SNE [79], UMAP [52], and other DR methods to reduce the high-dimensional word embeddings of each layer to 2D and visualize them in the form of scatter plots to understand the semantic information learned by the model through the change in distance between words.…”

Section: Interpretability Analysis Methods For Transformersmentioning

confidence: 99%

See 1 more Smart Citation

Visual Explanation for Open-Domain Question Answering With BERT

Zekai

Sun

Zhao

et al. 2024

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Interpretability Analysis Methods For Transformersmentioning

confidence: 99%

Visual Explanation for Open-Domain Question Answering With BERT

Zekai

Sun

Zhao

et al. 2024

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

show abstract

“…Others learn new distance functions for MDS to update the projection to best respect user manipulations [8,42]. Fujiwara et al provide a visual analytics framework for comparative analysis, providing interactions to manipulate and update projections to illustrate the similarities and differences between clusters of points [22]. This work expands on past work by specifically targeting imaged data to provide both projection-steering interactions and visual explanations of the 2D space.…”

Section: Semantic Interactionmentioning

confidence: 99%

Explainable Interactive Projections for Image Data

Han

Faust

Norambuena

et al. 2022

Advances in Visual Computing

View full text Add to dashboard Cite

Making sense of large collections of images is difficult. Dimension reductions (DR) assist by organizing images in a 2D space based on similarities, but provide little support for explaining why images were placed together or apart in the 2D space. Additionally, they do not provide support for modifying and updating the 2D space to explore new relationships and organizations of images. To address these problems, we present an interactive DR method for images that uses visual features extracted by a deep neural network to project the images into 2D space and provides visual explanations of image features that contributed to the 2D location. In addition, it allows people to directly manipulate the 2D projection space to define alternative relationships and explore subsequent projections of the images. With an iterative cycle of semantic interaction and explainable-AI feedback, people can explore complex visual relationships in image data. Our approach to human-AI interaction integrates visual knowledge from both human mental models and pre-trained deep neural models to explore image data. Two usage scenarios are provided to demonstrate that our method is able to capture human feedback and incorporate it into the model. Our visual explanations help bridge the gap between the feature space and the original images to illustrate the knowledge learned by the model, creating a synergy between human and machine that facilitates a more complete analysis experience. Firstly, I would like to thank my advisor Dr. Christopher North for his continued support in my research study. He was and remains my best role model for a teacher, mentor, and researcher.

show abstract

“…To find similarities and differences between groups of a dataset, Fujiwara et al [15] describe a technique to interactively adjust embeddings by moving or scaling group representations and showing the impact of attributes on the embeddings' axes, carrying on from previous work on interactive embeddings [16], [17]. Ma and Maciejewski [18] describe the analysis of class separations in embeddings through locally linear segments, which connects the work to other recent efforts to explain non-linear embeddings [19], [20], [21].…”

Section: Exploration Of Embedding Spacesmentioning

confidence: 99%

Visual Exploration of Relationships and Structure in Low-Dimensional Embeddings

Eckelt

Hinterreiter

Adelberger

et al. 2023

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

In this work, we propose an interactive visual approach for the exploration and formation of structural relationships in embeddings of high-dimensional data. These structural relationships, such as item sequences, associations of items with groups, and hierarchies between groups of items, are defining properties of many real-world datasets. Nevertheless, most existing methods for the visual exploration of embeddings treat these structures as second-class citizens or do not take them into account at all. In our proposed analysis workflow, users explore enriched scatterplots of the embedding, in which relationships between items and/or groups are visually highlighted. The original high-dimensional data for single items, groups of items, or differences between connected items and groups is accessible through additional summary visualizations. We carefully tailored these summary and difference visualizations to the various data types and semantic contexts. During their exploratory analysis, users can externalize their insights by setting up additional groups and relationships between items and/or groups. We demonstrate the utility and potential impact of our approach by means of two use cases and multiple examples from various domains.

show abstract

Interactive Dimensionality Reduction for Comparative Analysis

Cited by 27 publications

References 71 publications

Visual Explanation for Open-Domain Question Answering With BERT

Visual Explanation for Open-Domain Question Answering With BERT

Explainable Interactive Projections for Image Data

Visual Exploration of Relationships and Structure in Low-Dimensional Embeddings

Contact Info

Product

Resources

About