TECHNICAL POINT OF CONTACT (TPOC)
MARK ZAPPAVIGNA
SEAN SISTI
PROJECTED CMMC LEVEL REQUIREMENT
Level 2 (Self)
TECHNOLOGY AREAS
Air Platform
|
Battlespace
|
Human Systems
|
Information Systems
|
Sensors
|
Space Platforms
|
Weapons
MODERNIZATION PRIORITIES
Advanced Computing and Software
|
Integrated Sensing and Cyber
|
Microelectronics
|
Quantum Sciences
|
Space Technology
|
Trusted AI and Autonomy
KEYWORDS
Neural Architecture Integration, Brain-Inspired AI, Heterogeneous Computing, Cognitive Architectures, Integrated Neural Networks, Neocortex-Inspired Architectures, Cognitive Computing
OBJECTIVE
This SBIR solicitation seeks innovative research and development of novel AI architectures that integrate multiple neural network paradigms into a unified framework. To develop novel hardware and software architectures that integrate multiple neural network modalities into a single system, inspired by the functional organization of the human neocortex. This research will explore innovative approaches to achieve low SWAP (Size, Weight, and Power) while enhancing computational efficiency and cognitive capabilities relevant to DoD applications.The objective of this research is to develop integrated neural architectures that:Combine multiple neural network models (e.g., convolutional neural networks, recurrent neural networks, attention-based mechanisms) into a single architecture.Emulate the emergent properties of cognitive processing in the neocortex, such as hierarchical organization and processing, contextual dependence and attention, memory and working memory, and decision-making and uncertainty estimation.Achieve low SWAP (Size, Weight, and Power) to enable deployment on resource-constrained platforms (e.g., wearable devices, drones, autonomous vehicles).Demonstrate improved performance and robustness over state-of-the-art architectures on various cognitive tasks (e.g., object recognition, natural language processing, spatial reasoning).
DESCRIPTION
Current artificial intelligence (AI) systems predominantly rely on single, monolithic neural network architectures, which limits their ability to match the human brain's remarkable computational efficiency and adaptability. The human neocortex achieves this through its intricate interplay of diverse neuronal populations and specialized cortical areas. This SBIR solicitation invites proposals for research that aims to emulate this complex organization by integrating multiple neural architectures such as convolutional networks, recurrent networks, transformers, spiking neural networks, and biologically-inspired models within a single architecture.This research seeks to bridge the gap between current AI limitations and the flexible, multi-modal processing capabilities observed in the human brain. Proposals should focus on developing innovative methodologies for: (1) seamlessly interconnecting diverse neural modules, (2) dynamically allocating computational resources based on task demands, and (3) achieving significant reductions in Size, Weight, and Power (SWAP) consumption compared to traditional AI architectures.This integration should leverage principles inspired by the neocortex's hierarchical organization and functional specialization, aiming to achieve:Enhanced Cognitive Capabilities: Develop AI systems capable of performing complex tasks requiring multi-modal perception, reasoning, and decision-making, surpassing the limitations of single-architecture approaches.Improved Efficiency and Adaptability: Explore novel hardware and software solutions that enable efficient information flow and communication between integrated networks, leading to improved computational efficiency and adaptability to novel situations.This research has the potential to advance the field of artificial intelligence by creating more efficient, robust, and brain-inspired AI systems with applications in fields like robotics, healthcare, and autonomous systems.
PHASE I
Conduct a thorough review of existing neural network architectures and their integration strategies. Develop a conceptual design for an integrated architecture inspired by neocortical principles. Demonstrate the feasibility of key architectural components through simulations or small-scale prototypes. Define potential low SWAP hardware solutions (e.g., neuromorphic computing, spiking neural networks).
PHASE II
Develop a fully functional prototype system implementing the integrated architecture. Conduct extensive testing and evaluation of the prototype against relevant DoD tasks and benchmarks. Optimize hardware and software components for minimal SWAP while maintaining performance.
PHASE III DUAL USE APPLICATIONS
The Phase III effort targets autonomous systems for military use and commercial applications in robotics and healthcare. Entering at TRL 6, the integrated brain-inspired AI system will be optimized for tactical air dominance and operationally focused ABMS. Transition planning involves coordination with AFRL/RI and potential engagements with DAF customers for further development and fielding.
REFERENCES
Pan, W., Zhao, F., Zhao, Z., & Zeng, Y. (2024). Brain-inspired evolutionary architectures for spiking neural networks. IEEE Transactions on Artificial Intelligence, 5(11), 5760-5770.
Schmidgall, S., Ziaei, R., Achterberg, J., Kirsch, L., Hajiseyedrazi, S., & Eshraghian, J. (2024). Brain-inspired learning in artificial neural networks: a review. APL Machine Learning, 2(2).
de Jong, J., Voelker, A. R., Stewart, T. C., Aky rek, E. G., Eliasmith, C., & van Rijn, H. (2025). A unified neurocomputational model of prospective and retrospective timing. Psychological Review.
Eliasmith, C., Stewart, T. C., Choo, X., Bekolay, T., DeWolf, T., Tang, Y., & Rasmussen, D. (2012). A large-scale model of the functioning brain. science, 338(6111), 1202-1205.
QUESTIONS & ANSWERS
NOTE:
To ask a question, you must
log in or create an account
for the DSIP.
