Back to results
US 2026/0290440 A1Published application
Artificial intelligence processor and method of processing deep-learning operation using the same
AssigneeElectronics and Telecommunications Research Institute
InventorsJin-Ho HAN, Byung-Jo KIM, Ju-Yeob KIM, Hye-Ji KIM, Joo-Hyun LEE, Seong-Min KIM
Life of this patent
Estimating the term…
Abstract
Disclosed herein is an Artificial Intelligence (AI) processor. The AI processor includes multiple NVM AI cores for respectively performing basic unit operations required for a deep-learning operation based on data stored in NVM; SRAM for storing at least some of the results of the basic unit operations; and an AI core for performing an accumulation operation on the results of the basic unit operation.
Claims18
An Artificial Intelligence (AI) processing system, comprising:
- a Non-volatile Memory (NVM) array;
- one or more first operators for respectively performing primary operations required for a deep-learning operation based on data stored in the NVM array;
- on-chip memory for storing at least part of results of the primary operations;
- and
- a second operator for performing a subsequent operation using the results of the primary operations.
2 more elements
A method for processing a deep-learning operation, comprising:
- performing, by respective one or more first operators, primary operations required for a deep-learning operation based on data stored in a Non-volatile Memory (NVM) array;
- storing at least part of results of the primary operations in on-chip memory;
- and
- performing, by a second operator, a subsequent operation using the results of the primary operations.
1 more element
Dependencies are inferred from the claim text.