# US 2023/0267337 A1 · Conditional noise layers for generating adversarial examples

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

- Number: US 2023/0267337 A1
- Document ID: US20230267337A1
- Status: Published application
- Kind: A1
- Application number: 18/114,165
- Filed: 2023-02-24
- Published: 2023-08-24
- Earliest claimed: 2022-02-24
- Assignees: Protopia AI, Inc.
- Applicants: Protopia AI, Inc.
- Inventors: Hadi Esmaeilzadeh; Anwesa Choudhuri
- Web page: https://patentgrep.com/patent/US20230267337A1

## CPC

- G06N 3/094: Adversarial learning
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/047: Probabilistic or stochastic networks
- G06N 3/0895: Weakly supervised learning, e.g. semi-supervised or self-supervised learning
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Provided is a process including: obtaining, with a computer system, a data set having labeled members with labels designating corresponding members as belonging to corresponding classes; training, with the computer system, a machine learning model having deterministic layers and a parallel set of conditional layers each corresponding to a different class among the corresponding classes, wherein training includes adjusting parameters of the machine learning model according to an objective function that is differentiable; and storing, with the computer system, the trained machine learning model in memory.

## Claims

### Claim 1

A non-transitory computer-readable storage medium storing instructions that when executed by one or more processors perform operations comprising:
obtaining, with a computer system, a data set having labeled members with labels designating corresponding members as belonging to corresponding classes;
training, with the computer system, a machine learning model having deterministic layers and a parallel set of conditional layers each corresponding to a different class among the corresponding classes, wherein training includes adjusting parameters of the machine learning model according to an objective function that is differentiable; and
storing, with the computer system, the trained machine learning model in memory.

### Claim 2 (depends on 1)

The medium of claim 1, wherein the parallel set of conditional layers are stochastic layers and wherein training includes learning, for at least one parameter in each of the parallel set of stochastic layers, a corresponding distribution to be randomly sampled from during operation of the machine learning model.

### Claim 3 (depends on 2)

The medium of claim 2, wherein:
the respective distributions are parametric statistical distributions, each characterized, at least in part, by a respective pair of statistical parameters; and
the operations further comprise learning, using gradient descent, for each of the respective distributions, the respective pairs of statistical parameters based on an objective function, wherein the objective function is differentiable with respect to the respective pairs of statistical parameters of the respective probability distributions.

### Claim 4 (depends on 1)

The medium of claim 1, wherein the machine learning model further comprises a selection layer configured to select among the parallel set of layers based on a class of input data.

### Claim 5 (depends on 1)

The medium of claim 1, further comprising determining a measure of robustness of the machine learning model based on the trained parallel set of conditional layers.

### Claim 6 (depends on 5)

The medium of claim 5, wherein determining the measure of robustness comprises determining a magnitude based on the trained parallel set of conditional layers.

### Claim 7 (depends on 5)

The medium of claim 5, wherein determining a measure of robustness of the machine learning model comprises determining a measure of robustness for a given condition corresponding to a give of the parallel set of conditional layers.

### Claim 8 (depends on 1)

The medium of claim 1, the operations further comprising determining an adversarial example based on a given of the parallel set of conditional layers.

### Claim 9 (depends on 1)

The medium of claim 1, the operations further comprising generating a set of adversarial attack training data based on the parallel set of conditional layers.

### Claim 10 (depends on 9)

The medium of claim 9, the operations further comprising additionally training the machine learning model based on the set of adversarial attack training data.

### Claim 11 (depends on 1)

The medium of claim 1, wherein training according to the objective function includes adjusting parameters of the machine learning model to maximize noise in the parallel set of conditional layers while minimizing loss in the model.

### Claim 12 (depends on 1)

The medium of claim 1, wherein training according to the objective function includes adjusting parameters of the machine learning model to minimize noise in the parallel set of conditional layers while maximizing accuracy of the model.

### Claim 13 (depends on 1)

The medium of claim 1, wherein the operations comprise steps for learning distributions of the parallel set of conditional layers.

### Claim 14 (depends on 1)

The medium of claim 1, wherein the operations comprise steps for applying the parallel set of conditional layers to the machine learning model.

### Claim 15 (depends on 1)

The medium of claim 1, wherein the parallel set of conditional layers are convolutional layers.

### Claim 16 (depends on 1)

The medium of claim 1, wherein at least some of the labeled members of the training set are obfuscated during training.

### Claim 17 (depends on 1)

The medium of claim 1, further comprising obfuscating data based on the trained machine learning model.

### Claim 18 (depends on 1)

The medium of claim 1, wherein the machine learning model further comprises a regularization layer.

### Claim 19 (depends on 1)

The medium of claim 1, wherein the machine learning model is a neural network.

### Claim 20

A method comprising:
obtaining, with a computer system, a data set having labeled members with labels designating corresponding members as belonging to corresponding classes;
training, with the computer system, a machine learning model having deterministic layers and a parallel set of conditional layers each corresponding to a different class among the corresponding classes, wherein training includes adjusting parameters of the machine learning model according to an objective function that is differentiable; and
storing, with the computer system, the trained machine learning model in memory.
