WizMap: Scalable Interactive Visualization for Exploring Large Machine Learning Embeddings

crown jewel figure
WizMap empowers machine learning researchers and domain experts to easily explore and interpret millions of embedding vectors across different levels of granularity. Consider the task of investigating the embeddings of all 63k natural language processing paper abstracts indexed in ACL Anthology from 1980 to 2022. (A) The Map View tightly integrates a contour layer, a scatter plot, and automatically-generated multi-resolution embedding summaries to help users navigate through the large embedding space. (B) The Search Panel enables users to rapidly test their hypotheses through fast full-text embedding search. (C) The Control Panel allows users to customize embedding visualizations, compare multiple embedding groups and observe how embeddings evolve over time.
Demo Video
Machine learning models often learn latent embedding representations that capture the domain semantics of their training data. These embedding representations are valuable for interpreting trained models, building new models, and analyzing new datasets. However, interpreting and using embeddings can be challenging due to their opaqueness, high dimensionality, and the large size of modern datasets. To tackle these challenges, we present WizMap, an interactive visualization tool to help researchers and practitioners easily explore large embeddings. With a novel multi-resolution embedding summarization method and a familiar map-like interaction design, WizMap enables users to navigate and interpret embedding spaces with ease. Leveraging modern web technologies such as WebGL and Web Workers, WizMap scales to millions of embedding points directly in users' web browsers and computational notebooks without the need for dedicated backend servers. WizMap is open-source and available at the following public demo link: https://poloclub.github.io/wizmap.
WizMap: Scalable Interactive Visualization for Exploring Large Machine Learning Embeddings
  title = {{{WizMap}}: {{Scalable}} Interactive Visualization for Exploring Large Machine Learning Embeddings},
  booktitle = {Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 3: {{System}} Demonstrations)},
  author = {Wang, Zijie J. and Hohman, Fred and Chau, Duen Horng},
  year = {2023},
  url = {https://aclanthology.org/2023.acl-demo.50}