公开(公告)号：US11928584B2

公开(公告)日：2024-03-12

申请号：US16778587

申请日：2020-01-31

Applicant: salesforce.com, inc.

Inventor： Bradford William Powley ,  Noah Burbank ,  Rowan Cassius

IPC: G06N3/08 ,  H04L67/1001

CPC classification number: G06N3/08 ,  H04L67/1001

Abstract: Methods, systems, and devices for distributed hyperparameter tuning and load balancing are described. A device (e.g., an application server) may generate a first set of combinations of hyperparameter values associated with training a mathematical model. The mathematical model may include a machine learning model, an optimization model, or any combination. The device may identify a subset of combinations from the first set of combinations that are associated with a computational runtime that exceeds a first threshold and may distribute the subset of combinations across a set of machines. The device may then test each of the first set of combinations in a parallel processing operation to generate a first set of validation error values and may test a second set of combinations of hyperparameter values using an objective function that is based on the first set of validation error values.

公开(公告)号：US20210241164A1

公开(公告)日：2021-08-05

申请号：US16778587

申请日：2020-01-31

Applicant: salesforce.com, Inc.

Inventor： Bradford William Powley ,  Noah William Burbank ,  Rowan Cassius

IPC: G06N20/00 ,  G06F30/27 ,  H04L29/08

Abstract: Methods, systems, and devices for distributed hyperparameter tuning and load balancing are described. A device (e.g., an application server) may generate a first set of combinations of hyperparameter values associated with training a mathematical model. The mathematical model may include a machine learning model, an optimization model, or any combination. The device may identify a subset of combinations from the first set of combinations that are associated with a computational runtime that exceeds a first threshold and may distribute the subset of combinations across a set of machines. The device may then test each of the first set of combinations in a parallel processing operation to generate a first set of validation error values and may test a second set of combinations of hyperparameter values using an objective function that is based on the first set of validation error values.