The U.S. Army has created a tool that addresses cyber and physical challenges affecting private data sharing between fielded soldiers.
The service branch said Thursday its Army Research Laboratory worked with partners to develop the DoppelGANger tool that models and assesses time series data based on network and system datasets.
Carnegie Mellon University and IBM worked with ARL to study how generative adversarial networks can support data sharing through a generic framework where synthetic datasets can be shared with minimal technical skill.
“We identify key challenges of existing GAN approaches for such workloads with respect to fidelity, such as long-term dependencies, complex multidimensional relationships and mode collapse and privacy, as existing guarantees are poorly understood and can sacrifice fidelity,” said Zinan Lin, a doctoral student at CMU.
The National Science Foundation, Google, JP Morgan Chase and Siemens compose ARL's partners to support the effort.
The effort's researchers will go on to further identify and expand DoppelGANger's applications. The study has so far found the tool's potential for use with medical, financial, speech, music and text data.
