Invention Grant
- Patent Title: Linearly augmented neural network
-
Application No.: US14980976Application Date: 2015-12-28
-
Publication No.: US10558909B2Publication Date: 2020-02-11
- Inventor: James G. Droppo , Pegah Ghahremani , Michael L. Seltzer
- Applicant: Microsoft Technology Licensing, LLC
- Applicant Address: US WA Redmond
- Assignee: Microsoft Technology Licensing, LLC
- Current Assignee: Microsoft Technology Licensing, LLC
- Current Assignee Address: US WA Redmond
- Agency: Kelly, Holt & Christensen, PLLC
- Agent Christopher J. Volkmann
- Main IPC: G06N3/04
- IPC: G06N3/04 ; G06N3/08
Abstract:
A neural network is structured to connect the input values of an input set, at each level, to that level's output using a linear bypass connection. The linear bypass connection passes the input values, to the output, without applying a non-linear function to them.
Public/Granted literature
- US20170185887A1 LINEARLY AUGMENTED NEURAL NETWORK Public/Granted day:2017-06-29
Information query
