Paypal, Inc.

Artum Zolotushko

Shlomi Boutnaru

Yuri Shafet

Eyal Ben Simon

Amit Benbassat

Techniques for identifying weaknesses in a probabilistic model such as an artificial neural network using an iterative process are disclosed. A seed file may be obtained and variant files generated therefrom. The variant files may be evaluated for their fitness, based upon the ability of the variant files to cause the probabilistic model to fail. The fittest variants, which may refer to those variants that are most successful in causing the model to fail, may be selected. From these selected variants, a next generation of variant files may be created. The next generation of variant files may be evaluated for their fitness. At each step of fitness evaluation or at the end of the iterative process, a map of the fittest variants may be generated to identify hotspots. These hotspots may reveal segments of code or a file that are problematic for the model, which can be used to improve the model.

Hotspots for probabilistic model testing and cyber analysis

Systems and methods for integrating, into a first software program binary, segments of a second software program are disclosed. The integration causes the execution of segments of the second software program as the first software program binary is executed. In one embodiment, a second software program, such as an embeddable software application, is received and divided into a plurality of segments, each segment corresponding to a portion of the embeddable software application. Instrumentation points corresponding to the segments of the embeddable software application are inserted into a plurality of locations within a software binary to create a modified software binary. The modified software binary thus includes the selected software binary and the embeddable software program.

Growing community of inventors

Pardess-Hanna, Israel

Artum Zolotushko