Invention Application
- Patent Title: SKIMMING DATA SEQUENCES USING RECURRENT NEURAL NETWORKS
-
Application No.: US16865747Application Date: 2020-05-04
-
Publication No.: US20200265191A1Publication Date: 2020-08-20
- Inventor: Quoc V. Le , Hongrae Lee , Wei Yu
- Applicant: Google LLC
- Main IPC: G06F40/284
- IPC: G06F40/284 ; G06N3/08 ; G06N3/04

Abstract:
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing sequential data. In one aspect, a computer-implemented method includes receiving a request to generate a system output for an input data sequence, the input data sequence including a plurality of tokens. One or more tokens may be designated as tokens to be skipped. When a token has not been designated as a token to be skipped, the token is processed using a recurrent neural network to update a current internal state of the recurrent neural network. The system output is generated from the final internal state of the recurrent neural network.
Public/Granted literature
- US11048875B2 Skimming data sequences using recurrent neural networks Public/Granted day:2021-06-29
Information query