Invention Grant
- Patent Title: Effective chip yield for artificial intelligence integrated circuit with embedded memory
-
Application No.: US16230020Application Date: 2018-12-21
-
Publication No.: US10733039B2Publication Date: 2020-08-04
- Inventor: Chyu-Jiuh Torng , Daniel H. Liu , Wenhan Zhang , Hualiang Yu
- Applicant: Gyrfalcon Technology Inc.
- Applicant Address: US CA Milpitas
- Assignee: Gyrfalcon Technology Inc.
- Current Assignee: Gyrfalcon Technology Inc.
- Current Assignee Address: US CA Milpitas
- Agency: Brinks Gilson & Lione
- Main IPC: G06F11/00
- IPC: G06F11/00 ; G06F11/07 ; G11C29/18 ; G11C29/44 ; G06N3/063

Abstract:
This disclosure relates to testing of integrated artificial intelligence (AI) circuit with embedded memory to improve effective chip yield and to mapping addressable memory segments of the embedded memory to multilayer AI networks at the network level, layer level, parameter level, and bit level based on bit error rate (BER) of the addressable memory segments. The disclosed methods and systems allows for deployment of one or more multilayer AI networks in an AI circuit with sufficient model accuracy even when the embedded memory has an overall BER higher than a preferred overall threshold.
Information query