Systems and methods for secure authentication based on machine learning techniques
A system described herein may provide a technique for the use of machine learning techniques to perform authentication, such as biometrics-based user authentication. For example, user biometric information (e.g., facial features, fingerprints, voice, etc.) of a user may be used to train a machine learning model, in addition to a noise vector. A representation of the biometric information (e.g., an image file including a picture of the user's face, an encoded file with vectors or other representation of the user's fingerprint, a sound file including the user's voice, etc.) may be iteratively transformed until the transformed biometric information matches the noise vector, and the machine learning model may be trained based on the set of transformations that ultimately yield the noise vector, when given the biometric information.
1. A device, comprising:
one or more processors configured to:
apply a particular set of transformations to a first set of biometric data;
determine that a first measure of similarity between a second set of data and the first set of biometric data, with the particular set of transformations applied, exceeds a first threshold measure of similarity, wherein the second set of data includes a noise vector;
maintain, in a storage associated with the device and based on determining that the first measure of similarity exceeds the first threshold measure of similarity, the particular set of transformations and the second set of data, without maintaining the first set of biometric data in the storage associated with the device;
apply, based on determining that the first measure of similarity exceeds the first threshold measure of similarity, the particular set of transformations to a third set of data;
determine a second measure of similarity, between the second set of data and the third set of data with the particular set of transformations applied; and
determine, based on whether the second measure of similarity exceeds a second threshold measure of similarity, whether the third set of data matches the first set of biometric data without comparing the third set of data to the first set of biometric data.
2. The device of claim 1 , wherein the third set of data includes data received as part of an authentication request, wherein the one or more processors are further configured to:
accept the authentication request when determining that the first set of biometric data matches the third set of data, and
deny the authentication request when determining that the first set of biometric data does not match the third set of data.
3. The device of claim 2 , wherein the third set of data includes biometric information received as part of the authentication request.
4. The device of claim 2 , wherein the first set of biometric data is received as part of a registration of an authentication mechanism with which the authentication request is associated.
5. The device of claim 1 , wherein the first set of biometric data includes at least one of:
image data,
fingerprint data, or
audio data.
6. The device of claim 1 , wherein the particular set of transformations is a first set of transformations, wherein the one or more processors are further configured to:
apply a second set of transformations to the first set of biometric data;
determine that a third measure of similarity between the second set of data and the first set of biometric data, with the second set of transformations applied, does not exceed the threshold measure of similarity; and
select the first set of transformations to apply to the first set of biometric data based on determining that the third measure of similarity, between the second set of data and the first set of biometric data with the second set of transformations applied, does not exceed the threshold measure of similarity,
wherein applying the first set of transformations is performed after determining that the third measure of similarity, between the second set of data and the first set of biometric data with the second set of transformations applied, does not exceed the threshold measure of similarity.
7. A method, comprising:
applying a particular set of transformations to a first set of data;
determining that a first measure of similarity between a second set of data and the first set of biometric data, with the particular set of transformations applied, exceeds a first threshold measure of similarity, wherein the second set of data includes a noise vector;
maintaining, in a storage associated with the device and based on determining that the first measure of similarity exceeds the first threshold measure of similarity, the particular set of transformations and the second set of data, without maintaining the first set of biometric data in the storage associated with the device;
applying, based on determining that the first measure of similarity exceeds the first threshold measure of similarity, the particular set of transformations to a third set of data;
determining a second measure of similarity, between the second set of data and the third set of data with the particular set of transformations applied; and
determining, based on whether the second measure of similarity exceeds a second threshold measure of similarity, whether the third set of data matches the first set of biometric data without comparing the third set of data to the first set of biometric data.
8. A non-transitory computer-readable medium, storing a plurality of processor-executable instructions to:
apply a particular set of transformations to a first set of data;
determine that a first measure of similarity between a second set of data and the first set of biometric data, with the particular set of transformations applied, exceeds a first threshold measure of similarity, wherein the second set of data includes a noise vector;
maintain, in a storage associated with the device and based on determining that the first measure of similarity exceeds the first threshold measure of similarity, the particular set of transformations and the second set of data, without maintaining the first set of biometric data in the storage associated with the device;
apply, based on determining that the first measure of similarity exceeds the first threshold measure of similarity, the particular set of transformations to a third set of data;
determine a second measure of similarity, between the second set of data and the third set of data with the particular set of transformations applied; and
determine, based on whether the second measure of similarity exceeds a second threshold measure of similarity, whether the third set of data matches the first set of biometric data without comparing the third set of data to the first set of biometric data.
9. The non-transitory computer-readable medium of claim 8 , wherein the third set of data includes data received as part of an authentication request, wherein the plurality of processor-executable instructions further include processor-executable instructions to:
accept the authentication request when determining that the first set of biometric data matches the third set of data, and
deny the authentication request when determining that the first set of biometric data does not match the third set of data.
10. The non-transitory computer-readable medium of claim 9 , wherein the third set of data includes biometric information received as part of the authentication request.
11. The non-transitory computer-readable medium of claim 9 , wherein the first set of biometric data is received as part of a registration of an authentication mechanism with which the authentication request is associated.
12. The non-transitory computer-readable medium of claim 8 , wherein the first set of biometric data includes at least one of:
image data,
fingerprint data, or
audio data.
13. The non-transitory computer-readable medium of claim 8 , wherein the particular set of transformations is a first set of transformations, wherein the plurality of processor-executable instructions further include processor-executable instructions to:
apply a second set of transformations to the first set of biometric data;
determine that a third measure of similarity between the second set of data and the first set of biometric data, with the second set of transformations applied, does not exceed the threshold measure of similarity; and
select the first set of transformations to apply to the first set of biometric data based on determining that the third measure of similarity, between the second set of data and the first set of biometric data with the second set of transformations applied, does not exceed the threshold measure of similarity,
wherein applying the first set of transformations is performed after determining that the third measure of similarity, between the second set of data and the first set of biometric data with the second set of transformations applied, does not exceed the threshold measure of similarity.
14. The device of claim 1 , wherein the first set of data includes at least one of:
image data,
fingerprint data, or
voice data.
15. The method of claim 7 , wherein the first set of data includes at least one of:
image data,
fingerprint data, or
voice data.
16. The method of claim 7 , wherein the third set of data includes data received as part of an authentication request, the method further comprising:
accepting the authentication request when determining that the first set of biometric data matches the third set of data, and
denying the authentication request when determining that the first set of biometric data does not match the third set of data.
17. The method of claim 16 , wherein the third set of data includes biometric information received as part of the authentication request.
18. The method of claim 16 , wherein the first set of biometric data is received as part of a registration of an authentication mechanism with which the authentication request is associated.
19. The method of claim 7 , wherein the particular set of transformations is a first set of transformations, the method further comprising:
applying a second set of transformations to the first set of biometric data;
determining that a third measure of similarity between the second set of data and the first set of biometric data, with the second set of transformations applied, does not exceed the threshold measure of similarity; and
selecting the first set of transformations to apply to the first set of biometric data based on determining that the third measure of similarity, between the second set of data and the first set of biometric data with the second set of transformations applied, does not exceed the threshold measure of similarity,
wherein applying the first set of transformations is performed after determining that the third measure of similarity, between the second set of data and the first set of biometric data with the second set of transformations applied, does not exceed the threshold measure of similarity.
20. The non-transitory computer-readable medium of claim 8 , wherein the first set of data includes at least one of:
image data,
fingerprint data, or
voice data.