Invention Grant
- Patent Title: Parsimonious inference on convolutional neural networks
-
Application No.: US15815171Application Date: 2017-11-16
-
Publication No.: US11321613B2Publication Date: 2022-05-03
- Inventor: Ilias Theodorakopoulos , Vassileios Pothos , Dimitris Kastaniotis , Nikos Fragoulis
- Applicant: IRIDA LABS S.A.
- Applicant Address: GR Patras
- Assignee: IRIDA LABS S.A.
- Current Assignee: IRIDA LABS S.A.
- Current Assignee Address: GR Patras
- Agency: Sheridan Ross, PC
- Agent Jason H. Vick
- Main IPC: G06N3/08
- IPC: G06N3/08 ; G06N3/04

Abstract:
The disclosed system incorporates a new learning module, the Learning Kernel Activation Module (LKAM), at least serving the purpose of enforcing the utilization of less convolutional kernels by learning kernel activation rules and by actually controlling the engagement of various computing elements: The exemplary module activates/deactivates a sub-set of filtering kernels, groups of kernels, or groups of full connected neurons, during the inference phase, on-the-fly for every input image depending on the input image content and the learned activation rules.
Public/Granted literature
- US20180137417A1 PARSIMONIOUS INFERENCE ON CONVOLUTIONAL NEURAL NETWORKS Public/Granted day:2018-05-17
Information query