Defense Advanced Research Projects AgencyOur Research

Our Research

DARPA’s investment strategy begins with a portfolio approach. Reaching for outsized impact means taking on risk, and high risk in pursuit of high payoff is a hallmark of DARPA’s programs. We pursue our objectives through hundreds of programs. By design, programs are finite in duration while creating lasting revolutionary change. They address a wide range of technology opportunities and national security challenges. This assures that while individual efforts might fail—a natural consequence of taking on risk—the total portfolio delivers. More

For reference, past DARPA research programs can be viewed in the Past Programs Archive.

The past decade has seen explosive growth in development and training of artificial intelligence (AI) systems. However, as AI has taken on progressively more complex problems, the amount of computation required to train the largest AI systems has been increasing ten-fold annually. While AI advances are beginning to have a deep impact in digital computing processes, trade-offs between computational capability, resources and size, weight, and power consumption (SWaP) will become increasingly critical in the near future. More
Efficient discovery and production of new molecules is essential to realize capabilities across the DoD, from simulants and medicines essential to counter emerging threats, to coatings, dyes and specialty fuels needed for advanced performance. More
The United States Government has an interest in developing and maintaining a strategic understanding of events, situations, and trends around the world, in a variety of domains. The information used in developing this understanding comes from many disparate sources, in a variety of genres, and data types, and as a mixture of structured and unstructured data. Unstructured data can include text or speech in English and a variety of other languages, as well as images, videos, and other sensor information. More
The Artificial Intelligence Research Associate (AIRA) program is part of a broad DAPRA initiative to develop and apply “Third Wave” AI technologies that are robust to sparse data and adversarial spoofing, and that incorporate domain-relevant knowledge through generative contextual and explanatory models. More
Humans intuitively combine pre-existing knowledge with observations and contextual clues to construct rich mental models of the world around them and use these models to evaluate goals, perform thought experiments, make predictions, and update their situational understanding. When the environment contains other people, humans use a skill called theory of mind (ToM) to infer their mental states from observed actions and context, and predict future actions from those inferred states. More
The Automating Scientific Knowledge Extraction (ASKE) program aims to develop technology to automate some of the manual processes of scientific knowledge discovery, curation and application. ASKE is part of DARPA's Artificial Intelligence Exploration (AIE) program, a key component of the agency’s broader AI investment strategy aimed at ensuring the United States maintains an advantage in this critical and rapidly accelerating technology area. More
Some of the systems that matter most to the Defense Department are very complicated. Ecosystems, brains and economic and social systems have many parts and processes, but they are studied piecewise, and their literatures and data are fragmented, distributed and inconsistent. It is difficult to build complete, explanatory models of complicated systems, and so effects in these systems that are brought about by many interacting factors are poorly understood. More
| AI | Automation | Data |
The Bioelectronics for Tissue Regeneration (BETR) program will develop technology aimed at speeding warfighter recovery, and thus resilience, by directly intervening in wound healing. To do this, researchers will build an adaptive system that uses actuators to biochemically or biophysically stimulate tissue, sensors to track the body’s complex response to that stimulation, and adaptive learning algorithms to integrate sensor data and dictate intervention to the actuators. More
An emergent type of geopolitical warfare in recent years has been coined "gray zone competition," or simply "competition," because it sits in a nebulous area between peace and conventional conflict. It’s not openly declared or defined, it’s slower and is prosecuted more subtly using social, psychological, religious, information, cyber and other means to achieve physical or cognitive objectives with or without violence. The lack of clarity of intent in competition activity makes it challenging to detect, characterize, and counter an enemy fighting this way. More
The Communicating with Computers (CwC) program aims to enable symmetric communication between people and computers in which machines are not merely receivers of instructions but collaborators, able to harness a full range of natural modes including language, gesture and facial or other expressions. For the purposes of the CwC program, communication is understood to be the sharing of complex ideas in collaborative contexts. Complex ideas are assumed to be built from a relatively small set of elementary ideas, and language is thought to specify such complex ideas—but not completely, because language is ambiguous and depends in part on context, which can augment language and improve the specification of complex ideas. More
| AI | Autonomy | Data |
In order to transform machine learning systems from tools into partners, users need to trust their machine counterpart. One component to building a trusted relationship is knowledge of a partner’s competence (an accurate insight into a partner’s skills, experience, and reliability in dynamic environments). While state-of-the-art machine learning systems can perform well when their behaviors are applied in contexts similar to their learning experiences, they are unable to communicate their task strategies, the completeness of their training relative to a given task, the factors that may influence their actions, or their likelihood to succeed under specific conditions. More
The DARPA Space Environment Exploitation (SEE) program seeks to develop new models and sensing modalities to predict and observe the dynamics of the near-earth space environment. The SEE program explores how to go beyond magnetohydrodynamic descriptions of the magnetosphere, ionosphere, thermosphere coupled system to include wave/wave, wave/particle, and particle/particle interactions while using the latest advances in high performance computing such as GPUs and TPUs. More
Department of Defense (DoD) operators and analysts collect and process copious amounts of data from a wide range of sources to create and assess plans and execute missions. However, depending on context, much of the information that could support DoD missions may be implicit rather than explicitly expressed. Having the capability to automatically extract operationally relevant information that is only referenced indirectly would greatly assist analysts in efficiently processing data. More
Deep Purple aims to advance the modeling of complex dynamic systems using new information-efficient approaches that make optimal use of data and known physics at multiple scales. The program is investigating next-generation deep learning approaches that use not only high throughput multimodal scientific data from observations and controlled experiments (including behaviors such as phase transitions and chaos), but also of the known science of such systems at whatever scales it exists. More
The Digital RF Battlespace Emulator (DRBE) program aims to create the world’s first, large-scale, virtual RF environment for developing, training, and testing advanced radio frequency (RF) systems. The DRBE system will seek to enable numerous RF systems such as radar and electronic warfare (EW) systems to interact with each other in a fully closed-loop RF environment. More
Dramatic success in machine learning has led to a torrent of Artificial Intelligence (AI) applications. Continued advances promise to produce autonomous systems that will perceive, learn, decide, and act on their own. However, the effectiveness of these systems is limited by the machine’s current inability to explain their decisions and actions to human users (Figure 1). The Department of Defense (DoD) is facing challenges that demand more intelligent, autonomous, and symbiotic systems. Explainable AI—especially explainable machine learning—will be essential if future warfighters are to understand, appropriately trust, and effectively manage an emerging generation of artificially intelligent machine partners. More
The Von Neumann architecture has significantly aided the rapid advancement of computing over the past seven decades. However, moving data between the processors and memory components of this architecture requires significant time and high-energy consumption, which constrains the computing performance and workload. Overcoming this bottleneck requires new computing architectures and devices that can significantly advance the computing performance beyond the traditional practice of transistor scaling (i.e., Moore’s Law). More
The goal of the Fundamental Design (FUN Design) program is to determine whether we can develop or discover a new set of building blocks to describe conceptual designs. The design building blocks will capture the components’ underlying physics allowing a family of nonintuitive solutions to be generated. More
Rapid comprehension of world events is essential for informing U.S. national security - a task that becomes more difficult as the amount of unstructured, multimedia information grows exponentially. Humans make sense of events by organizing them into narrative structures that occur frequently. These structures are abstracted into schemas, which are organized units of knowledge that represent a pattern of memory used in human cognition. More
| AI | Analytics | Data |
In supervised machine learning (ML), the ML system learns by example to recognize things, such as objects in images or speech. Humans provide these examples to ML systems during their training in the form of labeled data. With enough labeled data, we can generally build accurate pattern recognition models. More
| AI | Algorithms | Data |
Artificial intelligence (AI) and machine learning (ML) systems have advanced significantly in recent years. Despite a wide range of impressive results, current AI is not intelligent in the biological sense. These systems are limited to performing only those tasks for which they have been specifically programmed and trained, and are inherently subject to safety hazards when encountering situations outside them. More
The U.S. Government operates globally and frequently encounters so-called “low-resource” languages for which no automated human language technology capability exists. Historically, development of technology for automated exploitation of foreign language materials has required protracted effort and a large data investment. Current methods can require multiple years and tens of millions of dollars per language—mostly to construct translated or transcribed corpora. More
Machine common sense has long been a critical—but missing—component of AI. Its absence is perhaps the most significant barrier between the narrowly focused AI applications we have today and the more general, human-like AI systems we would like to build in the future. The MCS program seeks to create the computing foundations needed to develop machine commonsense services to enable AI applications to understand new situations, monitor the reasonableness of their actions, communicate more effectively with people, and transfer learning to new domains. More
The Physics of Artificial Intelligence (PAI) program is part of a broad DAPRA initiative to develop and apply “Third Wave” AI technologies to sparse data and adversarial spoofing, and that incorporate domain-relevant knowledge through generative contextual and explanatory models. More
Machine learning – the ability of computers to understand data, manage results and infer insights from uncertain information – is the force behind many recent revolutions in computing. Email spam filters, smartphone personal assistants and self-driving vehicles are all based on research advances in machine learning. Unfortunately, even as the demand for these capabilities is accelerating, every new application requires a Herculean effort. Teams of hard-to-find experts must build expensive, custom tools that are often painfully slow and can perform unpredictably against large, complex data sets. More
Researchers have demonstrated effective attacks on machine learning (ML) algorithms. These attacks can cause high-confidence misclassifications of input data, even if the attacker lacks detailed knowledge of the ML classifier algorithm and/or training data. Developing effective defenses against such attacks is essential if ML is to be used for defense, security, or health and safety applications. More
The goal of the Radio Frequency Machine Learning Systems (RFMLS) Program is to develop the foundations for applying modern data-driven Machine Learning (ML) to the RF Spectrum domain. These innovations form the basis of a new wave of Signal Processing technologies to address performance limitations of conventionally designed radio frequency (RF) systems such as radar, signals intelligence, electronic warfare, and communications. More
Current artificial intelligence (AI) systems excel at tasks defined by rigid rules – such as mastering the board games Go and chess with proficiency surpassing world-class human players. However, AI systems aren’t very good at adapting to constantly changing conditions commonly faced by troops in the real world – from reacting to an adversary’s surprise actions, to fluctuating weather, to operating in unfamiliar terrain. More
Serial Interactions in Imperfect Information Games Applied to Complex Military Decision Making (SI3-CMD) builds on recent developments in artificial intelligence and game theory to enable more effective decisions in adversarial domains. SI3-CMD will explore several military decision making applications at strategic, tactical, and operational levels and develop AI/game theory techniques appropriate for their problem characteristics. More
Cyber physical systems (CPS) are instrumental to current and future Department of Defense (DoD) mission needs – unmanned vehicles, weapon systems, and mission platforms are all examples of military-relevant CPS. These systems and platforms integrate cyber and physical subsystems, and the enormous complexity of the resulting CPS has made their engineering design a daunting challenge. An immediate consequence of this complexity is development cycles with prolonged timelines that challenge DoD’s ability to counter emerging threats. More
New manufacturing technologies such as additive manufacturing have vastly improved the ability to create shapes and material properties previously thought impossible. Generating new designs that fully exploit these properties, however, has proven extremely challenging. Conventional design technologies, representations, and algorithms are inherently constrained by outdated presumptions about material properties and manufacturing methods. As a result, today’s design technologies are simply not able to bring to fruition the enormous level of physical detail and complexity made possible with cutting-edge manufacturing capabilities and materials. More
The Understanding Group Biases (UGB) program seeks to develop and prove out capabilities that can radically enhance the scale, speed, and scope of automated, ethnographic-like methods for capturing group biases and cultural models from increasingly available large digital datasets. More
Successful integration of next generation AI into DoD applications must be able to deal with incomplete, sparse and noisy data as well as unexpected circumstances that might arise while solving real world problems. Thus, there is a need for new computing models that are efficient and robust, can learn new concepts with very few examples, and can guide the development of adequate novel hardware to support them. More
| AI | Algorithms | Math |
Currently, understanding and assessing the readiness of the warfighter involves medical intervention with the help of advanced equipment, such as electrocardiographs (EKGs) and other specialized medical devices, that are too expensive and cumbersome to employ continuously or without supervision in non-controlled environments. On the other hand, currently 92 percent of adults in the United States own a cell phone, which could be used as the basis for continuous, passive health, and readiness assessment. More
| AI | Analytics | Data | Health |
The World Modelers program aims to develop technology that integrates qualitative causal analyses with quantitative models and relevant data to provide a comprehensive understanding of complicated, dynamic national security questions. The goal is to develop approaches that can accommodate and integrate dozens of contributing models connected by thousands of pathways—orders of magnitude beyond what is possible today. More
| AI | Automation | Data |