March 2, 2020
Conference Paper

Parallel Embeddings: a Visualization Technique for Contrasting Learned Representations

Abstract

We introduce ``Parallel Embeddings'', a new technique that generalizes the classical Parallel Coordinates visualization technique from a sequence of features within a single matrix to a sequence of multiple feature matrices. This is accomplished by separately embedding each matrix in the sequence into a 2-D frame, juxtaposing these 2-D representations, then connecting points between frames. We outline how this visualization can be used for several ``model comparison'' tasks in machine learning. We compare user performance with Parallel Embeddings to a baseline using the TensorFlow Embedding Projector for estimating model accuracy and understanding conceptual model differences. We found that users were more accurate with Parallel Embeddings for model comparison and that users learned how to use Parallel Embeddings more quickly than TensorFlow Embedding Projector. Furthermore, users' analytical process using Embedding Projector was positively affected by using Parallel Embeddings beforehand.

Revised: March 18, 2020 | Published: March 2, 2020

Citation

Arendt D.L., N. Nur, Z. Huang, G. Fair, and W. Dou. 2020. Parallel Embeddings: a Visualization Technique for Contrasting Learned Representations. In Proceedings of the 25th International Conference on Intelligent User Interfaces (IUI 2020), March 17-20, 2020, Cagliari, Itally, 259–274. New York, New York:Association for Computing Machinery. PNNL-SA-148252. doi:10.1145/3377325.3377514