# US 2023/0409982 A1 · Artificial neural network emulation of hotspots

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0409982 A1
- Document ID: US20230409982A1
- Status: Published application
- Kind: A1
- Application number: 18/456,057
- Filed: 2023-08-25
- Published: 2023-12-21
- Earliest claimed: 2019-11-25
- Assignees: Advanced Micro Devices, Inc.
- Applicants: Advanced Micro Devices, Inc.
- Inventors: Nicholas Malaya
- Web page: https://patentgrep.com/patent/US20230409982A1

## CPC

- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 3/096: Transfer learning
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Methods, devices, and systems for emulating a compute kernel with an ANN. The compute kernel is executed on a processor, and it is determined whether the compute kernel is a hotspot kernel. If the compute kernel is a hotspot kernel, the compute kernel is emulated with an ANN, and the ANN is substituted for the compute kernel.

## Claims

### Claim 1

A method for emulating a compute kernel with an artificial neural network (ANN), the method comprising:
executing the compute kernel on a processor; and
substituting an ANN for the compute kernel.

### Claim 2 (depends on 1)

The method of claim 1, wherein the ANN is substituted for the compute kernel responsive to the compute kernel comprising a hotspot kernel.

### Claim 3 (depends on 1)

The method of claim 1, wherein the ANN is substituted for the compute kernel responsive to a compute resource utilization of the compute kernel exceeding a compute resource utilization of a different compute kernel, or exceeding a threshold.

### Claim 4 (depends on 1)

The method of claim 1, further comprising emulating the compute kernel with the ANN, wherein the ANN comprises an emulation of the compute kernel, and wherein emulating the compute kernel with the ANN comprises offline-training the ANN prior to executing the compute kernel.

### Claim 5 (depends on 4)

The method of claim 4, wherein offline-training the ANN comprises:
inputting, to the ANN, training data typical of inputs to the compute kernel, comparing outputs from the ANN to known correct outputs corresponding to the training data; and adjusting the ANN based on differences between the outputs from the ANN and the known correct outputs.

### Claim 6 (depends on 1)

The method of claim 1, further comprising emulating the compute kernel with the ANN, wherein the ANN comprises an emulation of the compute kernel, and wherein emulating the compute kernel with the ANN comprises online-training the ANN based on execution of the compute kernel.

### Claim 7 (depends on 6)

The method of claim 6, wherein offline-training the ANN comprises:
inputting, to the ANN, inputs which were input to the compute kernel, comparing outputs from the ANN to known correct outputs corresponding to the inputs which were input to the compute kernel; and adjusting the ANN based on differences between the outputs from the ANN and the known correct outputs.

### Claim 8 (depends on 1)

The method of claim 1, further comprising emulating the compute kernel with the ANN, wherein the ANN comprises an emulation of the compute kernel, and wherein emulating the compute kernel with the ANN comprises offline-training the ANN prior to executing the compute kernel, and refining the ANN using online-training based on execution of the compute kernel.

### Claim 9 (depends on 1)

The method of claim 1, wherein substituting the ANN for the compute kernel comprises executing the ANN on the processor.

### Claim 10 (depends on 1)

The method of claim 1, wherein substituting the ANN for the compute kernel comprises executing the ANN on a different processor in communication with the processor.

### Claim 11

A computing device configured to emulate a compute kernel with an artificial neural network (ANN), the computing device comprising:
a processor configured to execute the compute kernel; and
the processor further configured to substitute an ANN for the compute kernel.

### Claim 12 (depends on 11)

The computing device of claim 11, wherein the processor is configured to substitute the ANN for the compute kernel responsive to the compute kernel comprising a hotspot kernel.

### Claim 13 (depends on 11)

The computing device of claim 11, wherein the processor is configured to substitute the ANN for the compute kernel responsive to a compute resource utilization of the compute kernel exceeding a compute resource utilization of a different compute kernel, or exceeding a threshold.

### Claim 14 (depends on 11)

The computing device of claim 11, wherein the ANN comprises an emulation of the compute kernel, and wherein emulating the compute kernel with the ANN comprises offline-training the ANN prior to executing the compute kernel.

### Claim 15 (depends on 14)

The computing device of claim 14, wherein offline-training the ANN comprises:
inputting, to the ANN, training data typical of inputs to the compute kernel, comparing outputs from the ANN to known correct outputs corresponding to the training data; and
adjusting the ANN based on differences between the outputs from the ANN and the known correct outputs.

### Claim 16 (depends on 11)

The computing device of claim 11, wherein the ANN comprises an emulation of the compute kernel, and wherein emulating the compute kernel with the ANN comprises online-training the ANN based on execution of the compute kernel.

### Claim 17 (depends on 16)

The computing device of claim 16, wherein the processor is further configured to offline-train the ANN, the processor further configured to input, to the ANN, inputs which were input to the compute kernel, compare outputs from the ANN to known correct outputs corresponding to the inputs which were input to the compute kernel; and adjust the ANN based on differences between the outputs from the ANN and the known correct outputs.

### Claim 18 (depends on 11)

The computing device of claim 11, wherein the ANN comprises an emulation of the compute kernel, and wherein emulating the compute kernel with the ANN comprises offline-training the ANN prior to executing the compute kernel, and refining the ANN using online-training based on execution of the compute kernel.

### Claim 19 (depends on 11)

The computing device of claim 11, wherein substituting the ANN for the compute kernel comprises executing the ANN on the processor.

### Claim 20 (depends on 11)

The computing device of claim 11, wherein substituting the ANN for the compute kernel comprises executing the ANN on a different processor in communication with the processor.
