ActiVis: Visual Exploration of Industry-Scale Deep Neural Network Models

• Published in: IEEE transactions on visualization and computer graphics Vol. 24; no. 1; pp. 88 - 97
• Main Authors: Minsuk Kahng, Andrews, Pierre Y., Kalro, Aditya, Chau, Duen Horng
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.01.2018
• Subjects: Computational modeling; Data models; Data visualization; deep learning; Facebook; information visualization; Machine learning; Neurons; Visual analytics
• DOI: 10.1109/TVCG.2017.2744718
• ISSN: 1077-2626; 1941-0506

While deep learning models have achieved state-of-the-art accuracies for many prediction tasks, understanding these models remains a challenge. Despite the recent interest in developing visual tools to help users interpret deep learning models, the complexity and wide variety of models deployed in industry, and the large-scale datasets that they used, pose unique design challenges that are inadequately addressed by existing work. Through participatory design sessions with over 15 researchers and engineers at Facebook, we have developed, deployed, and iteratively improved ActiVis, an interactive visualization system for interpreting large-scale deep learning models and results. By tightly integrating multiple coordinated views, such as a computation graph overview of the model architecture, and a neuron activation view for pattern discovery and comparison, users can explore complex deep neural network models at both the instance-and subset-level. ActiVis has been deployed on Facebook's machine learning platform. We present case studies with Facebook researchers and engineers, and usage scenarios of how ActiVis may work with different models.