# US 11,122,078 B1 · Systems and methods for private authentication with helper networks

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

- Number: US 11,122,078 B1
- Document ID: US11122078B1
- Status: Granted
- Kind: B1
- Application number: 17/183,950
- Filed: 2021-02-24
- Granted: 2021-09-14
- Earliest claimed: 2020-08-14
- Assignees: Private Identity LLC
- Applicants: Private Identity LLC
- Inventors: Scott Edward Streit
- Web page: https://patentgrep.com/patent/US11122078B1

## CPC

- G06F 21/32: using biometric data, e.g. fingerprints, iris scans or voiceprints
- G06F 7/02: Comparing digital values (G06F7/06, {G06F7/22,} G06F7/38 take precedence)
- G06F 7/5443: {Sum of products (for applications thereof, see the relevant places, e.g. G06F17/10, H03H17/00)}
- G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/045: Combinations of networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06V 10/772: Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries
- G06V 10/774: Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
- G06V 10/82: using neural networks
- G06V 10/993: {Evaluation of the quality of the acquired pattern}
- G06V 40/12: Fingerprints or palmprints
- G06V 40/16: Human faces, e.g. facial parts, sketches or expressions
- G06V 40/40: Spoof detection, e.g. liveness detection
- G10L 17/18: Artificial neural networks; Connectionist approaches
- H04L 63/0227: {Filtering policies (mail message filtering H04L51/212)}
- H04L 63/0236: {Filtering by address, protocol, port number or service, e.g. IP-address or URL}
- H04L 63/0861: {using biometrical features, e.g. fingerprint, retina-scan (cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231)}
- H04L 63/1416: {Event detection, e.g. attack signature detection}
- H04L 63/1466: {Active attacks involving interception, injection, modification, spoofing of data unit addresses, e.g. hijacking, packet injection or TCP sequence number attacks}
- G06F 2207/4824: Neural networks
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Helper neural network can play a role in augmenting authentication services that are based on neural network architectures. For example, helper networks are configured to operate as a gateway on identification information used to identify users, enroll users, and/or construct authentication models (e.g., embedding and/or prediction networks). Assuming, that both good and bad identification information samples are taken as part of identification information capture, the helper networks operate to filter out bad identification information prior to training, which prevents, for example, identification information that is valid but poorly captured from impacting identification, training, and/or prediction using various neural networks. Additionally, helper networks can also identify and prevent presentation attacks or submission of spoofed identification information as part of processing and/or validation.

## Claims

### Claim 1

An authentication system for privacy-enabled authentication, the system comprising:
at least one processor operatively connected to a memory;
an authentication data gateway, executed by the at least one processor, configured to filter identification information used in enrollment, identification, or authentication functions of subsequent neural networks, the authentication data gateway comprising at least:
a first pre-trained validation helper network associated with identification information of a first type comprising voice identification information, wherein the first pre-trained validation helper network is configured to:
evaluate an unknown identification sample of the first type, responsive to input of the unknown information sample of the first type to the first pre-trained validation helper network, wherein the first pre-trained validation helper network is pre-trained on evaluation criteria that is independent of a subject of the identification information seeking to be enrolled, identified, or authenticated;
responsive to a determination that the unknown identification sample meets the evaluation criteria, validate the unknown information sample for use in subsequent enrollment, identification, or authentication;
responsive to a determination that the unknown identification sample fails the evaluation criteria, reject the unknown information sample for use in subsequent enrollment, identification, or authentication; and
generate at least a binary evaluation of the unknown identification information sample based on the determination of the evaluation criteria, wherein the at least the binary evaluation includes generation of an output probability by the first pre-trained validation helper network that the unknown identification information sample is valid or invalid.

### Claim 2 (depends on 1)

The system of claim 1, wherein the first pre-trained validation helper network is configured to identify bad information samples, wherein bad information samples reduce identification accuracy of the subsequent neural networks.

### Claim 3 (depends on 1)

The system of claim 1, wherein the first pre-trained validation helper network is configured to identify good information samples, wherein good identification samples improve identification accuracy of the subsequent neural networks.

### Claim 4 (depends on 1)

The system of claim 1, wherein the first pre-trained validation helper network is configured to identify spoofed information samples.

### Claim 5 (depends on 4)

The system of claim 4, wherein the spoofed information samples include improper submission of valid identification information or indirect capture of valid authentication information.

### Claim 6 (depends on 1)

The system of claim 1, wherein the authentication data gateway further comprises at least a second pre-trained validation helper network associated with a second identification information type; wherein the at least the second validation helper network is trained on evaluation criteria independent of a subject of the identification information seeking to be enrolled, identified, or authenticated, and is configured to generate at least a binary evaluation of an unknown identification information sample of the second identification information type as valid or invalid identification information.

### Claim 7 (depends on 6)

The system of claim 6, wherein the second pre-trained validation helper network is configured to:
process a video or image input as identification information; and
output a probability that the video or image input is invalid.

### Claim 8 (depends on 6)

The system of claim 6, wherein the second pre-trained validation helper network is configured to:
process a video or image input as identification information, and
output a probability that the video or image input is a presentation attack.

### Claim 9 (depends on 6)

The system of claim 6, wherein the authentication data gateway further comprises a pre-trained geometry helper network configured to:
process identification information of the second type,
accept as input unencrypted identification information of the second type, and
communicate processed identification information of the second type to the second pre-trained validation helper network.

### Claim 10

A computer implemented method for privacy-enabled authentication, the method comprising:
filtering, by at least one processor, identification information used in enrollment, identification, or authentication functions of subsequent neural networks;
instantiating, by the at least one processor, a first pre-trained validation helper network associated with identification information of a first type comprising voice identification information, wherein the first pre-trained validation helper network is pre-trained on evaluation criteria that is independent of a subject of the identification information seeking to be enrolled, identified, or authenticated;
evaluating, by the first pre-trained validation helper network, an unknown identification sample of the first type, responsive to input of the unknown information sample of the first type to the first pre-trained validation helper network;
in response to determining, by the first pre-trained validation helper network, the unknown identification sample meets the evaluation criteria validating the unknown information sample for use in subsequent enrollment, identification, or authentication, or in response to determining if the unknown identification sample fails the evaluation criteria, rejecting the unknown information sample for use in subsequent enrollment, identification, or authentication; and
generating, by the first pre-trained validation helper network, at least a binary evaluation of the unknown identification information sample based on the determination of the evaluation criteria, wherein generating the at least the binary evaluation includes generating an output probability by the first pre-trained validation helper network that the unknown identification information sample is valid or invalid.

### Claim 11 (depends on 10)

The method of claim 10, wherein the act of determining includes identifying, by the first pre-trained validation helper network, bad information samples, wherein bad information samples reduce identification accuracy of the subsequent neural networks.

### Claim 12 (depends on 10)

The method of claim 10, wherein the act of determining includes identifying, by the first pre-trained validation helper network, good information samples, wherein good identification samples improve identification accuracy of the subsequent neural networks.

### Claim 13 (depends on 10)

The method of claim 10, wherein the method further comprises identifying, by the first pre-trained validation helper network, spoofed information samples.

### Claim 14 (depends on 13)

The method of claim 13, wherein the spoofed information samples include improper submission of valid identification information or indirect capture of valid authentication information.

### Claim 15 (depends on 10)

The method of claim 10, wherein the method further comprises:
instantiating at least a second pre-trained validation helper network associated with a second identification information type; wherein the at least the second validation helper network is trained on evaluation criteria independent of a subject of the identification information seeking to be enrolled, identified, or authenticated, and
generating at least a binary evaluation of an unknown identification information sample of the second identification information type as valid or invalid identification information, wherein generating the at least the binary evaluation includes generating an output probability from the at least the second pre-trained validation helper network that the unknown identification information sample is valid or invalid.

### Claim 16 (depends on 15)

The method of claim 15, wherein the method further comprises:
processing, by the second pre-trained validation helper network, a video or image input as identification information; and
generating a probability that the video or image input is invalid.

### Claim 17 (depends on 15)

The method of claim 15, wherein the method further comprises:
processing, by the second pre-trained validation helper network, a video or image input as identification information, and
generating a probability that the video or image input is a presentation attack.

### Claim 18 (depends on 15)

The method of claim 15, wherein the method further comprises instantiating, by the at least one processor, a first pre-trained geometry helper network configured to process identification information of the second type, accept as input unencrypted identification information of the second type, and communicate processed identification information of the second type to the first pre-trained validation helper network.

### Claim 19

An authentication system for privacy-enabled authentication, the system comprising:
at least one processor operatively connected to a memory;
an authentication data gateway, executed by the at least one processor, configured to filter identification information used in enrollment or identification functions of subsequent neural networks, the authentication data gateway comprising at least:
a first pre-trained validation helper network associated with identification information of a first type comprising voice identification information, wherein the first pre-trained validation helper network is further configured to:
evaluate identification information of the first type, wherein the first pre-trained validation helper network is trained on plaintext instances of good, bad, and spoofed identification samples, wherein good identification samples improve identification accuracy of the subsequent neural networks and bad identification samples reduce identification accuracy of the subsequent neural networks;
responsive to input of an unknown identification sample to the first pre-trained validation helper network, identify any bad information samples, good information samples, and spoofed information samples, regardless of a subject of the unknown identification sample; and
validate good information samples for use by subsequent neural network and reject bad and spoofed information samples from further processing based on at least a binary evaluation of the information sample, wherein the at least the binary evaluation includes generation of an output probability by the first pre-trained validation helper network that the unknown identification information sample is good or bad.