The growing utility of artificial intelligence (AI) is attributed to the development of neural networks. These networks are a class of models that make predictions based on previously observed data. While the inferential power of neural networks is great, the ability to explain their results is difficult because the underlying model is automatically generated. The AI community commonly refers to neural networks as black boxes because the patterns they learn from the data are not easily understood. This project aims to improve the visibility of patterns that neural networks identify in data. Through an interactive web application, NVIZ affords the user the ability to make their own neural network and train it to identify patterns in any data they choose. The network is displayed in a unique visual aesthetic that aims to intuitively demonstrate the influence that some features have over output variables. Modern UI/UX tricks (e.g., visual cues, animations) were employed to create an environment that feels familiar to webpages the user might have interacted with in the past. While neural networks are far from understood, we hope that through visual and kinesthetic play, the user will learn patterns alongside neural networks for a richer understanding of their function.
Hoffman, Kevin, "NVIZ: Unraveling Neural Networks Through Visualization" (2023). Mathematics and Computer Science Presentations. 4.
Available to all.
Computer Sciences Commons, Data Science Commons, Mathematics Commons
Presented as part of the Ursinus College Celebration of Student Achievement (CoSA) held April 19, 2023.
The downloadable file contains a poster.