Password discovery system using a generative adversarial network

Inventors

Gross, Samuel • Datta, Kaushik

Abstract

Systems and methods for password discovery are provided. A system receives a first password data set comprising known passwords and applies a rule-set to the first data set to generate a second password data set comprising passwords that are believed to be likely to be human-generated. The system trains a generative adversarial network, for generating predicted passwords, using the second data set, for example by incentivizing the GAN to favor passwords in the second data set. The system applies the generative adversarial network to generate a third password data set comprising predicted passwords. The system compares the third password data set to a data corpus to identify a string in the data corpus determined to match one of the predicted passwords in the first plurality of predicted passwords. The identified string may thus be identified as a previously undiscovered password, which may be applied to unlock password-protected systems and/or to further improve password discovery systems.

Core Innovation

The invention relates to a password discovery system for identifying passwords in a data corpus. The system receives a first password data set comprising a plurality of known passwords and applies a rule-set to generate a second password data set that is used for subsequent modeling and generation.

The system trains a generative adversarial network for generating predicted passwords using the second password data set. During training, the system generates a predicted password, determines whether the generated predicted password is included in the second password data set, and in response modifies a loss function of the generative adversarial network.

After training, the system applies the generative adversarial network to generate a third password data set comprising a first plurality of predicted passwords. The system compares the third password data set to a data corpus to identify a string in the data corpus determined to match one of the predicted passwords, and augments the second password data set based on the identified string by applying an extrapolation algorithm to generate extrapolated strings and adding the extrapolated strings to the second password data set.

Claims Coverage

The independent claims define 5 inventive features centered on password discovery using rule-set transformation, GAN training, corpus comparison, and extrapolation-based augmentation.

The claim set centers on rule-set generation of a second password data set from known passwords, GAN training with loss-function modification triggered by inclusion in the second password data set, generation of predicted passwords, matching predicted passwords against a data corpus, and augmentation of the second password data set using extrapolation.

Stated Advantages

Documented Applications