Invention Publication
- Patent Title: TECHNIQUES FOR KNOWLEDGE DISTILLATION BASED MULTI-VENDOR SPLIT LEARNING FOR CROSS-NODE MACHINE LEARNING
-
Application No.: US18340732Application Date: 2023-06-23
-
Publication No.: US20240056151A1Publication Date: 2024-02-15
- Inventor: June NAMGOONG , Yiyue CHEN , Taesang YOO , Abdelrahman Mohamed Ahmed Mohamed IBRAHIM
- Applicant: QUALCOMM Incorporated
- Applicant Address: US CA San Diego
- Assignee: QUALCOMM Incorporated
- Current Assignee: QUALCOMM Incorporated
- Current Assignee Address: US CA San Diego
- Main IPC: H04B7/06
- IPC: H04B7/06
Abstract:
The techniques described herein utilize a machine learning algorithm to train the encoders from multiple UE vendors and a shared decoder from a gNB vendor in order to develop a universal gNB decoder that may be capable of decoding input from UEs from different UE vendors at comparable performance and overhead to different decoders that are specifically developed for each encoder.
Information query
