Description
The Defense Advanced Research Projects Agency (DARPA) is soliciting innovative research proposals of interest to the Information Innovation Office (I2O). I2O explores game-changing technologies in the fields of information science and technology to anticipate and create rapid shifts in the complex national security landscape. Conflict can occur in traditional domains such as land, sea, air, and space, and in emerging domains such as cyber and other types of irregular warfare. I2Os research portfolio is focused on anticipating new modes of warfare in these emerging areas and developing the concepts and tools necessary to provide a decisive information advantage for the U.S. and its allies. As much of the world-wide economy has moved into cyberspace, protecting and assuring information flows over these networks has become a priority. Most networks today rely on the successive discovery of vulnerabilities and deployment of patches to maintain security. Even after patching, new vulnerabilities are often introduced in successive releases, and may even be introduced by the patches themselves. The I2O defensive cyber portfolio is largely focused on changing this paradigm through a variety of methods such as heterogeneity, formal methods proofs, secure code generation, and automation. Exploration of offensive methods is essential to expand and inform defensive work. The I2O portfolio covers a broad space, investigating enterprise networks, secure communications, industrial systems, and purpose-built military systems. I2O is also pursuing information technologies to change the way we perceive and interact with our surroundings. Exponential improvements in computing power, network bandwidth and storage density combine with ever more pervasive sensing and measurement technologies to provide new and powerful ways to gain insight into the world. Essentially all human activities that can be measured, from mercantile to military, are being quantitatively re-examined in the context of this new, big data, capability. Early demonstrations have produced remarkable insights into human activities and enabled quantitative decision making. I2Os investment portfolio begins at the fundamental science level with programs investigating varied topics from the mathematical properties of graphs, to online correlation for societal unrest. The portfolio also addresses fundamental computational issues such as novel algorithm design, natural language processing, and architectures for efficient processing of streamed data. Closer to the customer, I2O works with the national security community on operational data to ensure continuous transition of tools as programs make progress. I2O is interested in submissions related to the general areas discussed above and to the following specific areas: secure cloud computing systems software code with mathematically provable security properties automated cyber security systems preserving security on untrustworthy computational infrastructure insider threat detection and response maintaining security and mission effectiveness of systems after attack, including detection, cleanup, and reconstitution computational tools and scalable algorithms advanced user interfaces pre-processing technology for data sets, including those that are corrupted, incomplete, or disaggregated processing and computational approaches (including, but not limited to, novel algorithm design, natural language processing, and architecture systems) for data sets that may be multimodal, realtime-streamed, or on a scale for which storage is infeasible algorithm development for analysis of dynamic, unlinked and scripted content tools to draw inferences, deduce relationships, make correlations or detect anomalies working solely from data sets that are weak proxies for the underlying quantities of interest fusion of data from varied modalities ranging from traditional overhead sensing to open source data automated language translation and understanding fundamental science and mathematics supporting data analytics including: o mathematical properties of graphs o online correlation for societal unrest o virtual and blended worlds o machine-to-machine internet-of-things tools to assist emerging technologies in: o virtuality and immersivity o ubiquitous sensing o persuasion and cognitive hacking o social multimedia o hypercomputing environment modeling and reconstruction crowd-sourced approaches to data processing modeling of human activity and proclivities analytics of economic systems technologies for scalable education and training electronic warfare and cyber systems biometrics and provenance online information validation and evidence collection resilient command and control systems, including secure information exchange assured information access privacy science and systems semantic analysis programming languages I2O seeks unconventional approaches that are outside the mainstream, challenge accepted assumptions, and have the potential to radically change established practice. Proposed research should investigate innovative approaches that enable revolutionary advances in science, devices, or systems. Specifically excluded is research that primarily results in evolutionary improvements to the existing state of the art. See attached "I2O OFFICE WIDE DARPA-BAA-14-39" for further details.