-
公开(公告)号:US11625605B2
公开(公告)日:2023-04-11
申请号:US16723608
申请日:2019-12-20
Applicant: Nvidia Corporation
Inventor: Jonathan Edward Barker , Christopher Thomas Cheng , Paul Martin Springer , Wojciech Jablonski
Abstract: Apparatuses, systems, and techniques to optimize kernel selection for performing a computation. In at least one embodiment, a neural network is trained and utilized to generate a list of kernels so that an (e.g., optimal) kernel may be identified. The neural network receives characteristics of the input matrices and determines relevancy scores for a list of possible kernels. Based on an ordered listing of kernels by relevant score, a kernel is selected from the list and utilized to perform the computation and provide the result.