Interactive visualizations of a convolutional neural network

US10048826B2 · kind B2 · utility

Assignees

• SAS INSTITUTE, INC. · US
• North Carolina State University · US

Inventors

• Samuel Paul Leeman-Munk · Cary, US
• Saratendu Sethi · Raleigh, US
• Christopher Graham Healey · Cary, US
• Shaoliang Nie · Raleigh, US
• Kalpesh Padia · Raleigh, US
• Ravinder Devarajan · Cary, US
• David James Caira · Chapel Hill, US
• Jordan Riley Benson · Ellerbe, US
• James Allen Cox · New Brighton, US
• Lawrence E. Lewis · Kirkland, US
• Mustafa Onur Kabul · Apex, US

Key dates

• Filing date: Oct 3, 2017
• Grant date: Aug 14, 2018
• Expiry date: Oct 3, 2037

Classification

• Technology area (CPC G): Physics
• CPC primary: G06F2203/04803
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Interactive visualizations of a convolutional neural network are provided. For example, a graphical user interface (GUI) can include a matrix having symbols indicating feature-map values that represent likelihoods of particular features being present or absent at various locations in an input to a convolutional neural network. Each column in the matrix can have feature-map values generated by convolving the input to the convolutional neural network with a respective filter for identifying a particular feature in the input. The GUI can detect, via an input device, an interaction indicating that that the columns in the matrix are to be combined into a particular number of groups. Based on the interaction, the columns can be clustered into the particular number of groups using a clustering method. The matrix in the GUI can then be updated to visually represent each respective group of columns as a single column of symbols within the matrix.