Artificial intelligence (AI) model dependency handling in heterogeneous computing platforms

Patent metadata

Field	Value
Publication number	US-12585449-B2
Application number	US-202218056287-A
Country	US
Kind code	B2
Filing date	Nov 17, 2022
Priority date	Nov 17, 2022
Publication date	Mar 24, 2026
Grant date	Mar 24, 2026

Abstract

Systems and methods for Artificial Intelligence (AI) model dependency handling in heterogenous computing platforms are described. In an illustrative, non-limiting embodiment, an Information Handling System (IHS) may include a heterogeneous computing platform comprising a plurality of devices and a memory coupled to the platform, where the memory includes a plurality of sets of firmware instructions that, upon execution by a respective device, enables the respective device to provide a corresponding firmware service, and where at least one of the devices operates as an orchestrator configured to: receive an indication of a relationship between a first AI model and a second AI model; and in response to an instruction to update the first AI model and in the absence of another instruction to update the second AI model, trigger installation of an update to the second AI model.

First claim

The invention claimed is:

1 . An Information Handling System (IHS), comprising:
a heterogeneous computing platform comprising a plurality of hardware devices; and a memory coupled to the heterogeneous computing platform, wherein the memory contains a plurality of sets of firmware instructions, wherein each of the sets of firmware instructions, upon execution by a respective device among the plurality of hardware devices, enables the respective device to provide a corresponding firmware service, and wherein at least one of the plurality of hardware devices operates as an orchestrator configured to:
receive an indication of a relationship between a first Artificial Intelligence (AI) model and a second AI model;
in response to an instruction to update the first AI model and in the absence of another instruction to update the second AI model, trigger installation of an update to the second AI model, wherein the update to the second AI model is performed, at least in part, based upon contextual or telemetry data; and
send a message to one or more firmware services executed by a subset of the plurality of hardware devices via one or more Application Programming Interfaces (APIs) without any involvement by any host Operating System (OS) to collect the context or telemetry data.

2 . The IHS of claim 1 , wherein the heterogeneous computing platform comprises: a System-On-Chip (SoC), a Field-Programmable Gate Array (FPGA), or an Application-Specific Integrated Circuit (ASIC).

3 . The IHS of claim 1 , wherein the orchestrator comprises at least one of: a sensing hub, an Embedded Controller (EC), or a Baseboard Management Controller (BMC).

4 . The IHS of claim 1 , wherein the indication comprises a sign that an input into the second AI model is based upon an output from the first AI model.

5 . The IHS of claim 4 , wherein the first AI model comprises a noise suppression model, and wherein the second AI model comprises a volume adjustment model.

6 . The IHS of claim 1 , wherein the orchestrator is further configured to, in response to a determination that an input of a third AI model is based upon an output of the second AI model and in the absence of yet another instruction to update the third AI model, trigger installation of an update to the third AI model.

7 . The IHS of claim 1 , wherein the first AI model is deployable by a first one of the plurality of hardware devices and the second AI model is deployable by a second one of the plurality of hardware devices.

8 . The IHS of claim 1 , wherein to update the second AI model, the orchestrator is further configured to receive and deploy a version of the second AI model that precedes a latest version of the second AI model.

9 . The IHS of claim 1 , wherein the indication is received from an application executed by a host OS.

10 . The IHS of claim 1 , wherein the context or telemetry data comprises a metric indicative of at least one of: a core utilization, a memory utilization, a network utilization, a battery utilization, or a peripheral device utilization.

11 . The IHS of claim 1 , wherein the context or telemetry data comprises a metric indicative of at least one of: a presence of the user, an engagement of the user, an IHS location, or an IHS posture.

12 . The IHS of claim 1 , wherein the context or telemetry data comprises an identification of an application in execution by the IHS in a foreground window.

13 . The IHS of claim 1 , wherein the orchestrator is further configured to receive a policy from an Information Technology Decision Maker (ITDM) or Original Equipment Manufacturer (OEM).

14 . The IHS of claim 13 , wherein the policy identifies a remote service from which to retrieve a driver package usable to install the update to the second AI model.

15 . The IHS of claim 13 , wherein the policy comprises one or more rules, and wherein each rule associates at least one of: (a) the first AI model or a type of the first AI model, with (b) the second AI model or a type of the second AI model.

16 . The IHS of claim 15 , wherein the orchestrator is further configured to enforce the one or more rules based, at least in part, upon a comparison between current context or telemetry data and previous context or telemetry data.

17 . A method, comprising:
selecting a policy; and transmitting the policy to an Information Handling System (IHS) over a network, wherein the IHS comprises a heterogeneous computing platform having a plurality of devices, and wherein an orchestrator among the plurality of devices is configured to:
receive an indication of a relationship between a first Artificial Intelligence (AI) model and a second AI model; and
based upon the policy, perform at least one of:
in response to an update to the first AI model, update the second AI model; or
in response to an update to the second AI model, update the first AI model;
wherein the IHS comprises a laptop, wherein the update to the second AI model is performed, at least in part, based upon contextual or telemetry data, wherein the context or telemetry data comprises a metric indicative of at least one of: an IHS physical configuration or an IHS posture, and wherein the IHS posture comprises a laptop posture, a stand posture, a tablet posture, a book posture, a landscape posture, or a portrait posture.

18 . The method of claim 17 , wherein the IHS physical configuration comprises a lid position or lid angle.

Assignees

• Dell Products Lp

Inventors

• Kondapi Srikanth
• Hamlin Daniel L
• Vichare Nikhil Manohar

Classifications

• G06F8/65 Primary

  Updates (security arrangements therefor G06F21/57) · CPC title