公开(公告)号：US20220101096A1

公开(公告)日：2022-03-31

申请号：US17549645

申请日：2021-12-13

Applicant: Intel Corporation

Inventor： Gadi Singer ,  Nagib Hakim ,  Phillip Howard ,  Daniel Korat ,  Vasudev Lal ,  Arden Ma ,  Erik Norden ,  Ze'ev Rivlin ,  Ana Paula Quirino Simoes ,  Oren Pereg ,  Moshe Wasserblat

IPC: G06N3/04 ,  G06N5/02 ,  G06N5/04 ,  G06K9/62 ,  G06F11/34

Abstract: Methods and apparatus for a knowledge-based deep learning refactoring model with tightly integrated functional nonparametric memory are disclosed. An example non-transitory computer readable medium comprises instructions that, when executed, cause a machine to at least estimate a first information extraction cost corresponding to retrieval of information from a local knowledge base, estimate a second information extraction cost corresponding retrieval of information from a remote knowledge base, select an information source based on the first and second estimated information extraction costs, query the selected information source, in response to determining that the selected information source was an external information source, store the queried information in the local knowledge base, organize the stored information in the local knowledge base, and return the queried information.