# US 2025/0094768 A1 · Multi-branch machine learning models for multi-domain and multi-task processing

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

- Number: US 2025/0094768 A1
- Document ID: US20250094768A1
- Status: Published application
- Kind: A1
- Application number: 18/468,481
- Filed: 2023-09-15
- Published: 2025-03-20
- Earliest claimed: 2023-09-15
- Assignees: QUALCOMM Incorporated
- Applicants: QUALCOMM Incorporated
- Inventors: Andrii SKLIAR; Babak EHTESHAMI BEJNORDI; Amelie Marie Estelle ROYER; Tijmen Pieter Frederik BLANKEVOORT; Dushyant MEHTA
- Web page: https://patentgrep.com/patent/US20250094768A1

## CPC

- G06N 3/045: Combinations of networks
- G06N 3/0895: Weakly supervised learning, e.g. semi-supervised or self-supervised learning
- G06N 3/0475: Generative networks
- G06N 3/088: Non-supervised learning, e.g. competitive learning
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Certain aspects of the present disclosure provide techniques and apparatus for training and inferencing using a multi-domain machine learning model. An example method generally includes extracting, using a first neural network block, a plurality of features associated with inputs in a multi-domain input data set. A confusion matrix is generated based on the extracted plurality of features. A plurality of clusters is identified from the confusion matrix. Each cluster in the plurality of clusters generally corresponds to one or more data domains in the multi-domain input data set. A first gating neural network is trained based on the multi-domain input data set and the identified plurality of clusters. A plurality of second neural network blocks is trained based on a division of the multi-domain input data set into data associated with each cluster of the plurality of clusters.

## Claims

### Claim 1

A processing system, comprising:
at least one memory having executable instructions stored thereon; and
one or more processors configured to execute the executable instructions to cause the processing system to:
extract, using a first neural network block, a plurality of first features from a received input;
identify, using a gating neural network and the plurality of first features, a second neural network block from a plurality of second neural network blocks to use in processing the received input, each neural network in the plurality of second neural network blocks configured to process data from a cluster of domains in a universe of domains; and
process the received input using the identified second neural network block from the plurality of second neural network blocks.

### Claim 2 (depends on 1)

The processing system of claim 1, wherein to process the received input using the identified second neural network block, the one or more processors are configured to cause the processing system to:
extract a plurality of second features from the first features using the identified second neural network block;
identify, using a gating neural network associated with the identified second neural network block, a third neural network from a plurality of third neural networks to use in processing the received input, each neural network in the plurality of third neural networks being trained to perform a specified task; and
process the received input using the identified third neural network from the plurality of third neural networks.

### Claim 3 (depends on 1)

The processing system of claim 1, wherein to extract the plurality of first features from the received input, the one or more processors are configured to cause the processing system to:
extract features usable by the gating neural network to identify the second neural network block from the plurality of second neural network blocks to process the received input, and
extract features usable by the second neural network block from the plurality of second neural network blocks to generate an inference from the received input.

### Claim 4 (depends on 3)

The processing system of claim 3, wherein the second neural network block comprises a feature extractor block and a task head configured to generate the inference based on features extracted by the feature extractor block.

### Claim 5 (depends on 1)

The processing system of claim 1, wherein the gating neural network comprises a clustering neural network trained to classify an input into one of a plurality of data domains based on similarities to data in different data domains in a multi-domain data set.

### Claim 6 (depends on 1)

The processing system of claim 1, wherein the first neural network block and the plurality of second neural network blocks comprise jointly trained neural network blocks.

### Claim 7 (depends on 1)

The processing system of claim 1, wherein the plurality of second neural network blocks comprise image processing neural networks trained to classify images from one of a plurality of image data domains.

### Claim 8 (depends on 7)

The processing system of claim 7, wherein the processor is further configured to cause the processing system to output an output of the image processing neural networks to an autonomous vehicle control system to trigger generation of one or more control signals to control an autonomous vehicle based on the output of the image processing neural networks.

### Claim 9 (depends on 1)

The processing system of claim 1, wherein the plurality of second neural networks comprise generative artificial intelligence models trained to generate responses to prompts in one of a plurality of task-specific domains, and.

### Claim 10 (depends on 1)

The processing system of claim 1, wherein the plurality of second neural network blocks comprise generative artificial intelligence models trained to generate responses to prompts in one of a plurality of data-specific domains.

### Claim 11

A processor-implemented method, comprising:
extracting, using a first neural network block, a plurality of first features from a received input;
identifying, using a gating neural network and the plurality of first features, a second neural network block from a plurality of second neural network blocks to use in processing the received input, each neural network in the plurality of second neural network blocks being trained to process data from a cluster of related domains in a universe of domains; and
processing the received input using the identified second neural network block from the plurality of second neural network blocks.

### Claim 12 (depends on 11)

The method of claim 11, wherein processing the received input using the identified second neural network block comprises:
extracting a plurality of second features from the first features using the identified second neural network block;
identifying, using a gating neural network associated with the identified second neural network block, a third neural network from a plurality of third neural networks to use in processing the received input, each neural network in the plurality of third neural networks being trained to perform a specified task; and
processing the received input using the identified third neural network from the plurality of third neural networks.

### Claim 13 (depends on 11)

The method of claim 11, wherein extracting the plurality of first features from the received input comprises:
extracting features usable by the gating neural network to identify the second neural network block from the plurality of second neural network blocks to process the received input, and
extracting features usable by the second neural network block from the plurality of second neural network blocks to generate an inference from the received input.

### Claim 14 (depends on 13)

The method of claim 13, wherein the second neural network block comprises a feature extractor block and a task head configured to generate the inference based on features extracted by the feature extractor block.

### Claim 15 (depends on 11)

The method of claim 11, wherein the gating neural network comprises a clustering neural network trained to classify an input into one of a plurality of data domains based on similarities to data in different data domains in a multi-domain data set.

### Claim 16 (depends on 11)

The method of claim 11, wherein the first neural network block and the plurality of second neural network blocks comprise jointly trained neural network blocks.

### Claim 17 (depends on 11)

The method of claim 11, wherein the plurality of second neural network blocks comprise image processing neural networks trained to classify images from one of a plurality of image data domains.

### Claim 18 (depends on 17)

The method of claim 17, further comprising outputting an output of the image processing neural networks to an autonomous vehicle control system to trigger generation of one or more control signals to control an autonomous vehicle based on the output of the image processing neural networks.

### Claim 19 (depends on 11)

The method of claim 11, wherein the plurality of second neural networks comprise generative artificial intelligence models trained to generate responses to prompts in one of a plurality of task-specific domains, and.

### Claim 20 (depends on 11)

The method of claim 11, wherein the plurality of second neural network blocks comprise generative artificial intelligence models trained to generate responses to prompts in one of a plurality of data-specific domains.

### Claim 21

A processing system, comprising:
at least one memory having executable instructions stored thereon; and
one or more processors configured to execute the executable instructions to cause the processing system to:
extract, using a first neural network block, a plurality of features associated with inputs in a multi-domain input data set;
generate a confusion matrix based on the extracted plurality of features;
identify a plurality of clusters from the confusion matrix, each cluster in the plurality of clusters corresponding to one or more data domains in the multi-domain input data set;
train a first gating neural network based on the multi-domain input data set and the identified plurality of clusters; and
train a plurality of second neural network blocks based on a division of the multi-domain input data set into data associated with each cluster of the plurality of clusters.

### Claim 22 (depends on 21)

The processing system of claim 21, wherein the first gating neural network comprises a neural network configured to output features associated with a received input to a neural network block selected from the plurality of second neural network blocks for processing.

### Claim 23 (depends on 21)

The processing system of claim 21, wherein to identify the plurality of clusters from the confusion matrix, the one or more processors are configured to cause the processing system to identify data domains in the multi-domain input data set having a clustering quality score above a threshold value.

### Claim 24 (depends on 23)

The processing system of claim 23, wherein the clustering quality score comprises a distance score associated with a sampling ratio of a first data domain and a second data domain from the multi-domain input data set, and wherein the distance score is based on:
a joint distribution of data in the first data domain and the second data domain, and
a product distribution of data in the first data domain and the second data domain.

### Claim 25 (depends on 21)

The processing system of claim 21, wherein to train the plurality of second neural network blocks, the one or more processors are configured to cause the processing system to:
train the first neural network block to extract a set of features from the multi-domain input data set including a first set of features associated with the first gating neural network and a second set of features associated with a downstream task performed by a second neural network block from the plurality of second neural network blocks; and
train the second neural network block based on the second features.

### Claim 26 (depends on 25)

The processing system of claim 25, wherein to train the first neural network block and the second neural network block, the one or more processors are configured to cause the processing system to train the first neural network block and the second neural network block based on a joint loss function including a loss calculated for the first gating neural network and a loss calculated for the second neural network block.

### Claim 27 (depends on 25)

The processing system of claim 25, wherein to train the first neural network block to extract the first features and the second features, the one or more processors are configured to cause the processing system to extract a plurality of features through one or more temporary feature extractors.

### Claim 28 (depends on 27)

The processing system of claim 27, wherein the first neural network block comprises a frozen feature extractor configured to generate the first set of features and a plurality of trained feature extractors configured to generate the second set of features, the plurality of trained feature extractors being initialized based on parameters associated with a first feature extractor from the one or more temporary feature extractors.

### Claim 29 (depends on 21)

The processing system of claim 21, wherein the one or more processors are further configured to cause the processing system to:
train a second gating neural network to route requests to one of a plurality of third neural network blocks, the second gating neural network being associated with a neural network from the plurality of second neural network blocks; and
train the plurality of third neural network blocks to perform one or more specified tasks based on data in the multi-domain input data set associated with a cluster associated with the neural network from the plurality of second neural network blocks.

### Claim 30

A processor-implemented method, comprising:
extracting, using a first neural network block, a plurality of features associated with inputs in a multi-domain input data set;
generating a confusion matrix based on the extracted plurality of features;
identifying a plurality of clusters from the confusion matrix, each cluster in the plurality of clusters corresponding to one or more data domains in the multi-domain input data set;
training a first gating neural network based on the multi-domain input data set and the identified plurality of clusters; and
training a plurality of second neural network blocks based on a division of the multi-domain input data set into data associated with each cluster of the plurality of clusters.
