# US 2022/0092644 A1 · Pruning field weights for content selection

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

- Number: US 2022/0092644 A1
- Document ID: US20220092644A1
- Status: Published application
- Kind: A1
- Application number: 17/028,162
- Filed: 2020-09-22
- Published: 2022-03-24
- Earliest claimed: 2020-09-22
- Assignees: Oath Inc.
- Applicants: Oath Inc.
- Inventors: Junwei Pan; Tian Zhou; Aaron Eliasib Flores
- Web page: https://patentgrep.com/patent/US20220092644A1

## CPC

- G06Q 30/0275: {Auctions}
- G06N 3/082: modifying the architecture, e.g. adding, deleting or silencing nodes or connections
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06F 16/90344: {by using string matching techniques}
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

One or more computing devices, systems, and/or methods are provided. A machine learning model may be trained using a plurality of sets of information. One or more pruning operations may be performed in association with the training to generate a machine learning model with a sparse set of field weights associated with feature fields associated with features of the plurality of sets of auction information. A request for content associated with a client device may be received. A set of features associated with the request for content may be determined. Positive signal probabilities associated with a plurality of content items may be determined using the machine learning model based upon field weights, of the machine learning model, associated with the set of features. A content item may be selected from the plurality of content items for presentation via the client device based upon the positive signal probabilities.

## Claims

### Claim 1

A method, comprising:
receiving a first bid request, wherein:
the first bid request is associated with a first request for content associated with a first client device; and
the first bid request is indicative of a first set of features comprising one or more first features associated with the first request for content;
submitting a first bid value associated with a first content item to a first auction module for participation in a first auction associated with the first request for content;
storing, in an auction information database, a first set of auction information associated with the first auction, wherein:
the first set of auction information is indicative of the first set of features; and
the auction information database comprises a plurality of sets of auction information, comprising the first set of auction information, associated with a plurality of auctions comprising the first auction;
training a machine learning model using the plurality of sets of auction information;
performing one or more pruning operations, in association with the training, to generate a first machine learning model with a sparse set of field weights associated with feature fields associated with features of the plurality of sets of auction information, wherein each field weight of the sparse set of field weights is associated with two feature fields;
receiving a second bid request, wherein:
the second bid request is associated with a second request for content associated with a second client device;
the second bid request is indicative of a second set of features comprising a first feature associated with a first feature field and a second feature associated with a second feature field; and
a first field weight, associated with the first feature field and the second feature field, is pruned via the one or more pruning operations;
determining, using the first machine learning model, a plurality of click probabilities associated with a plurality of content items based upon one or more first field weights, of the first machine learning model, associated with the second set of features, wherein a first click probability of the plurality of click probabilities is associated with a second content item of the plurality of content items and corresponds to a probability of receiving a selection of the second content item responsive to presenting the second content item via the second client device;
selecting, from the plurality of content items, the second content item for presentation via the second client device based upon the plurality of click probabilities; and
submitting a second bid value associated with the second content item to a second auction module for participation in a second auction associated with the second request for content.

### Claim 2 (depends on 1)

The method of claim 1, wherein:
the one or more pruning operations are performed in an iterative pruning process.

### Claim 3 (depends on 2)

The method of claim 2, wherein:
the training the machine learning model comprises performing one or more first training steps to generate a first plurality of field weights;
the performing the one or more pruning operations comprises performing a first iteration of the iterative pruning process by setting a first subset of field weights, of the first plurality of field weights, to zero to generate a second plurality of field weights having a first sparsity;
the training the machine learning model comprises performing one or more second training steps, using the second plurality of field weights, to generate a third plurality of field weights; and
the performing the one or more pruning operations comprises performing a second iteration of the iterative pruning process by setting a second subset of field weights, of the third plurality of field weights, to zero to generate a fourth plurality of field weights having a second sparsity.

### Claim 4 (depends on 3)

The method of claim 3, wherein:
iterations of the iterative pruning process, comprising the first iteration and the second iteration, are performed until a fifth plurality of field weights is generated having a third sparsity that meets a target sparsity; and
the training the machine learning model comprises performing one or more third training steps, using the fifth plurality of field weights, to generate the sparse set of field weights.

### Claim 5 (depends on 3)

The method of claim 3, wherein:
iterations of the iterative pruning process, comprising the first iteration and the second iteration, are performed until the sparse set of field weights is generated having a third sparsity that meets a target sparsity.

### Claim 6 (depends on 3)

The method of claim 3, wherein:
the setting the first subset of field weights to zero is performed based upon a determination that field weights of the first subset of field weights are lowest field weights of the first plurality of field weights; and
the setting the second subset of field weights to zero is performed based upon a determination that field weights of the second subset of field weights are lowest field weights of the third plurality of field weights.

### Claim 7 (depends on 1)

The method of claim 1, wherein:
the one or more pruning operations are performed after the training the machine learning model.

### Claim 8 (depends on 7)

The method of claim 7, wherein:
the training the machine learning model comprises generating a second machine learning model with a first plurality of field weights; and
the one or more pruning operations are performed by setting a first subset of field weights, of the first plurality of field weights, to zero to generate the sparse set of field weights.

### Claim 9 (depends on 1)

The method of claim 1, comprising:
determining the second bid value based upon the first click probability.

### Claim 10 (depends on 1)

The method of claim 1, comprising:
receiving a click indication indicative of a selection of the first content item via the first client device, wherein the first set of auction information comprises the click indication.

### Claim 11 (depends on 1)

The method of claim 1, wherein:
prior to the performing the one or more pruning operations, the first field weight is equal to a first value; and
the determining the plurality of click probabilities is not performed based upon the first value.

### Claim 12

A computing device comprising:
a processor; and
memory comprising processor-executable instructions that when executed by the processor cause performance of operations, the operations comprising:
receiving a first request for content associated with a first client device;
determining, based upon the first request for content, a first set of features associated with the first request for content;
selecting a first content item for presentation via the first client device;
storing, in an information database, a first set of information associated with the first request for content, wherein:
the first set of information is indicative of the first set of features; and
the information database comprises a plurality of sets of information, comprising the first set of information, associated with a plurality of requests for content comprising the first request for content;
training a machine learning model using the plurality of sets of information;
performing one or more pruning operations in association with the training to generate a first machine learning model with a sparse set of field weights associated with feature fields associated with features of the plurality of sets of auction information, wherein each field weight of the sparse set of field weights is associated with two feature fields;
receiving a second request for content associated with a second client device;
determining, based upon the second request for content, a second set of features associated with the second request for content, wherein:
the second set of features comprises a first feature associated with a first feature field and a second feature associated with a second feature field; and
a first field weight, associated with the first feature field and the second feature field, is pruned via the one or more pruning operations;
determining, using the first machine learning model, a plurality of positive signal probabilities associated with a plurality of content items based upon one or more first field weights, of the first machine learning model, associated with the second set of features, wherein a first positive signal probability of the plurality of positive signal probabilities is associated with a second content item of the plurality of content items and corresponds to a probability of receiving a positive signal responsive to presenting the second content item via the second client device;
selecting, from the plurality of content items, the second content item for presentation via the second client device based upon the plurality of positive signal probabilities; and
transmitting the second content item to the second client device.

### Claim 13 (depends on 12)

The computing device of claim 12, wherein:
the one or more pruning operations are performed in an iterative pruning process.

### Claim 14 (depends on 13)

The computing device of claim 13, wherein:
the training the machine learning model comprises performing one or more first training steps to generate a first plurality of field weights;
the performing the one or more pruning operations comprises performing a first iteration of the iterative pruning process by setting a first subset of field weights, of the first plurality of field weights, to zero to generate a second plurality of field weights having a first sparsity;
the training the machine learning model comprises performing one or more second training steps, using the second plurality of field weights, to generate a third plurality of field weights; and
the performing the one or more pruning operations comprises performing a second iteration of the iterative pruning process by setting a second subset of field weights, of the third plurality of field weights, to zero to generate a fourth plurality of field weights having a second sparsity.

### Claim 15 (depends on 14)

The computing device of claim 14, wherein:
iterations of the iterative pruning process, comprising the first iteration and the second iteration, are performed until a fifth plurality of field weights is generated having a third sparsity that meets a target sparsity; and
the training the machine learning model comprises performing one or more third training steps, using the fifth plurality of field weights, to generate the sparse set of field weights.

### Claim 16 (depends on 14)

The computing device of claim 14, wherein:
iterations of the iterative pruning process, comprising the first iteration and the second iteration, are performed until the sparse set of field weights is generated having a third sparsity that meets a target sparsity.

### Claim 17 (depends on 14)

The computing device of claim 14, wherein:
the setting the first subset of field weights to zero is performed based upon a determination that field weights of the first subset of field weights are lowest field weights of the first plurality of field weights; and
the setting the second subset of field weights to zero is performed based upon a determination that field weights of the second subset of field weights are lowest field weights of the third plurality of field weights

### Claim 18 (depends on 12)

The computing device of claim 12, wherein:
the one or more pruning operations are performed after the training the machine learning model.

### Claim 19 (depends on 12)

The computing device of claim 12, wherein:
prior to the performing the one or more pruning operations, the first field weight is equal to a first value; and
the determining the plurality of positive signal probabilities is not performed based upon the first value.

### Claim 20

A non-transitory machine readable medium having stored thereon processor-executable instructions that when executed cause performance of operations, the operations comprising:
receiving a first bid request, wherein:
the first bid request is associated with a first request for content associated with a first client device; and
the first bid request is indicative of a first set of features comprising one or more first features associated with the first request for content;
submitting a first bid value associated with a first content item to a first auction module for participation in a first auction associated with the first request for content;
storing, in an auction information database, a first set of auction information associated with the first auction, wherein:
the first set of auction information is indicative of the first set of features; and
the auction information database comprises a plurality of sets of auction information, comprising the first set of auction information, associated with a plurality of auctions comprising the first auction;
training a machine learning model using the plurality of sets of auction information;
performing one or more pruning operations, in association with the training, to generate a first machine learning model with a sparse set of field weights associated with feature fields associated with features of the plurality of sets of auction information, wherein each field weight of the sparse set of field weights is associated with two feature fields;
receiving a second bid request, wherein the second bid request is associated with a second request for content associated with a second client device;
determining, based upon the second bid request, a second set of features associated with the second request for content, wherein:
the second set of features comprises a first feature associated with a first feature field and a second feature associated with a second feature field; and
a first field weight, associated with the first feature field and the second feature field, is pruned via the one or more pruning operations;
determining, using the first machine learning model, a plurality of click probabilities associated with a plurality of content items based upon one or more first field weights, of the first machine learning model, associated with the second set of features, wherein a first click probability of the plurality of click probabilities is associated with a second content item of the plurality of content items and corresponds to a probability of receiving a selection of the second content item responsive to presenting the second content item via the second client device;
selecting, from the plurality of content items, the second content item for presentation via the second client device based upon the plurality of click probabilities; and
submitting a second bid value associated with the second content item to a second auction module for participation in a second auction associated with the second request for content.
