# US 2026/0093523 A1 · Processing parallelism for machine learning model training

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

- Number: US 2026/0093523 A1
- Document ID: US20260093523A1
- Status: Published application
- Kind: A1
- Application number: 18/901,890
- Filed: 2024-09-30
- Published: 2026-04-02
- Earliest claimed: 2024-09-30
- Assignees: Advanced Micro Devices, Inc.
- Applicants: Advanced Micro Devices, Inc.
- Inventors: Sumanth Gudaparthi; Yao Cui Fehlis; Karthik Ramu Sangaiah; Sonali Singh
- Web page: https://patentgrep.com/patent/US20260093523A1

## CPC

- G06F 9/4881: {Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues}
- G06N 3/045: Combinations of networks
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06T 1/20: Processor architectures; Processor configuration, e.g. pipelining
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A processing system schedules parallel training of different instances of a machine learning model (MLM) based on a number of microbatches associated with training the machine learning model. The number of microbatches, along with the time required to complete a forward and backward pass of the MLM per microbatch, indicates the position, in time, of one or more expected idle cycles of a processing unit during training of a first instance of the MLM. A scheduler of the processing system schedules a second instance of the MLM during the one or more expected idle cycles.

## Claims

### Claim 1

A method comprising:
determining a number of microbatches associated with a training pass of a machine learning model; and
scheduling concurrent training of a first instance of the machine learning model and a second instance of the machine learning model at a plurality of processing units based on the determined number of microbatches.

### Claim 2 (depends on 1)

The method of claim 1, wherein the training pass includes a plurality of forward passes of the machine learning model, and further comprising:
determining a first number of processing cycles for executing the plurality of forward passes; and
scheduling the concurrent training further based on the determined first number of processing cycles.

### Claim 3 (depends on 2)

The method of claim 2, wherein the training pass includes a plurality of backward passes of the machine learning model, and further comprising:
determining a second number of processing cycles for executing the plurality of backward passes; and
scheduling the concurrent training further based on the determined second number of processing cycles.

### Claim 4 (depends on 1)

The method of claim 1, wherein the determined number of microbatches indicates timing of a plurality of idle cycles associated with training of the first instance of the machine learning model.

### Claim 5 (depends on 4)

The method of claim 4, wherein scheduling comprises scheduling training of the second instance of the machine learning model during the plurality of idle cycles associated with training of the first instance of the machine learning model.

### Claim 6 (depends on 1)

The method of claim 1, wherein scheduling comprises interleaving at least one training cycle of the second instance of the machine learning model between instances of the first instance of the machine learning model.

### Claim 7 (depends on 1)

The method of claim 1, wherein a first processing unit of the plurality of processing units executes a first layer of the first instance of the machine learning model, and a second processing unit of the plurality of processing units executes a second layer of the first instance of the machine learning model.

### Claim 8 (depends on 7)

The method of claim 7, wherein the first processing unit executes a first layer of the second instance of the machine learning model corresponding to the first layer of the first instance of the machine learning model.

### Claim 9

A method, comprising:
determining, based on a number of microbatches associated with training a machine learning model, a number of idle cycles at first processing unit; and
scheduling training of a first instance of the machine learning model and a second instance of the machine learning model based on the determined number of microbatches.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
determining a first number of processing cycles for executing a forward pass of the machine learning model; and
wherein scheduling training comprises scheduling training based on the first number of processing cycles.

### Claim 11 (depends on 10)

The method of claim 10, further comprising:
determining a second number of processing cycles for executing a backward pass of the machine learning model; and
wherein scheduling training comprises scheduling training based on the second number of processing cycles.

### Claim 12 (depends on 9)

The method of claim 9, wherein scheduling comprises scheduling training of the second instance of the machine learning model during idle cycles associated with training the first instance of the machine learning model.

### Claim 13

A processing system, comprising:
a plurality of processing units; and
a scheduler configured to:
determine a number of microbatches associated with a training pass of a machine learning model; and
schedule concurrent training of a first instance of the machine learning model and a second instance of the machine learning model at the plurality of processing units based on the determined number of microbatches.

### Claim 14 (depends on 13)

The processing system of claim 13, wherein the training pass includes a plurality of forward passes of the machine learning model, and wherein the scheduler is configured to:
determine a first number of processing cycles for executing the plurality of forward passes; and
schedule the concurrent training further based on the determined first number of processing cycles.

### Claim 15 (depends on 14)

The processing system of claim 14, wherein the training pass includes a plurality of backward passes of the machine learning model, and wherein the scheduler is configured to:
determining a second number of processing cycles for executing the plurality of backward passes; and
scheduling the concurrent training further based on the determined second number of processing cycles.

### Claim 16 (depends on 13)

The processing system of claim 13, wherein the determined number of microbatches indicates timing of a plurality of idle cycles associated with training of the first instance of the machine learning model.

### Claim 17 (depends on 16)

The processing system of claim 16, wherein the scheduler is configured to schedule training of the second instance of the machine learning model during the plurality of idle cycles associated with training of the first instance of the machine learning model.

### Claim 18 (depends on 13)

The processing system of claim 13, wherein scheduling comprises interleaving at least one training cycle of the second instance of the machine learning model between instances of the first instance of the machine learning model.

### Claim 19 (depends on 13)

The processing system of claim 13, wherein a first processing unit of the plurality of processing units executes a first layer of the first instance of the machine learning model, and a second processing unit of the plurality of processing units executes a second layer of the first instance of the machine learning model.

### Claim 20 (depends on 19)

The processing system of claim 19, wherein the plurality of processing units comprise graphics processing units (GPUs).