Invention Grant
- Patent Title: Recurrent neural networks having a probabilistic state component and state machines extracted from the recurrent neural networks
-
Application No.: US16524224Application Date: 2019-07-29
-
Publication No.: US11694062B2Publication Date: 2023-07-04
- Inventor: Cheng Wang , Mathias Niepert
- Applicant: NEC Laboratories Europe GmbH
- Applicant Address: DE Heidelberg
- Assignee: NEC CORPORATION
- Current Assignee: NEC CORPORATION
- Current Assignee Address: JP Tokyo
- Agency: Leydig, Voit & Mayer, Ltd.
- Main IPC: G06N3/044
- IPC: G06N3/044 ; G06N3/08 ; G06N3/047

Abstract:
A computer-implemented method includes instantiating a neural network including a recurrent cell. The recurrent cell includes a probabilistic state component. The method further includes training the neural network with a sequence of data. In an embodiment, the method includes extracting a deterministic finite automaton from the trained recurrent neural network and classifying a sequence with the extracted automaton.
Information query