THEORINET: Transferable, Hierarchical, Expressive, Optimal, Robust, and Interpretable NETworks

THEORINET: Transferable, Hierarchical, Expressive, Optimal, Robust, and Interpretable NETworks

Supported by: National Science Foundation, Simmons Foundation

Transferable, Hierarchical, Expressive, Optimal, Robust, and Interpretable NETworks (THEORINET) is an NSF-Simons Research Collaboration on the Mathematical and Scientific Foundations of Deep Learning (MoDL) whose goal is to develop a mathematical, statistical and computational framework that helps explain the success of current network architectures, understand its pitfalls, and guide the design of novel architectures with guaranteed confidence, robustness, interpretability, optimality, and transferability. THEORINET will also create new undergraduate and graduate programs in the foundations of data science and organize a series of collaborative research events, including semester research programs and summer schools on the foundations of deep learning.

THEORINET

EnCORE: The institute for Emerging CORE Methods in Data Science

EnCORE: The institute for Emerging CORE Methods in Data Science

Supported by: National Science Foundation Harnessing the Data Revolution

The Institute for Emerging CORE Methods in Data Science, or EnCORE, is led by the University of California San Diego in collaboration with the University of California, Los Angeles; University of Pennsylvania; and The University of Texas at Austin. EnCORE brings together scientists from multiple disciplines such as statistics, mathematics, electrical engineering, theoretical computer science, machine learning and health science, among others. EnCORE’s team will focus on the four CORE pillars of data science: C for complexities of data, O for optimization, R for responsible learning, and E for education and engagement. The institute is fostering a plan for outreach and broadening participation by engaging students of diverse backgrounds at all levels, from K-12 to postdocs and junior faculty. The project aims to reach a wide demography of students by offering collaborative courses across its partner universities and a flexible co-mentorship plan for multidisciplinary research. To bring theoretical development into practice, EnCORE will work with industry partners and domain scientists and will forge strong connections with other NSF Harnessing the Data Revolution Institutes across the nation.

EnCORE

IoT4Ag: The Internet of Things for Precision Agriculture

IoT4Ag: The Internet of Things for Precision Agriculture

Supported by: National Science Foundation Engineering Research Center

IoT4Ag will create novel, integrated systems that capture the microclimate and spatially, temporally, and compositionally map heterogeneous stresses for early detection and intervention to better outcomes in agricultural crop production. The Center will create internet of things (IoT) technologies to optimize practices for every plant; from sensors, robotics, and energy and communication devices to data-driven models constrained by plant physiology, soil, weather, management practices, and socio-economics.

IoT4Ag

Bridging offline design and online adaptation in safe learning-enabled systems

Bridging offline design and online adaptation in safe learning-enabled systems

Supported by: NSF Safe Learning-Enabled Systems

Maintaining the safety of a learning-enabled system that navigates in an unknown environment is a major challenge owing to uncertainty in the environment, the system’s goals, and the system’s learning-enabled components. This project proposes a novel approach to mitigating these uncertainties. The project’s novelties are the development of a two-phase design and deployment process integrated into a tight feedback loop: (1) an offline design process aimed at synthesizing systems that are provably robust and resilient to known unknowns, and (2) an automated online safety monitoring phase, during which a deployed learning-enabled system seeks to detect, learn about, and adapt to unknown unknowns. The project’s impacts include: (i) a mathematical guarantee on the end-to-end safety of the design and deployment process described above for learning-enabled systems; (ii) methods that ensure safety with respect to known unknowns during the offline design stage, and safety with respect to unknown unknowns during deployment, when possible; and (iii) techniques that identify and learn about unknown unknowns, that is, novel sources of uncertainty, so that they can be integrated into the design of future systems.

NSF SLES

Distributed and Collaborative Intelligent Systems and Technology

Distributed and Collaborative Intelligent Systems and Technology

Supported by: ARL Collaborative Research Alliance DCIST

The Distributed and Collaborative Intelligent Systems and Technology Collaborative Research Alliance (CRA) will create Autonomous, Resilient, Cognitive, Heterogeneous Swarms that can enable humans to participate in wide range of missions in dynamically changing, harsh and contested environments. These include search and rescue of hostages, information gathering after terrorist attacks or natural disasters, and humanitarian missions. Swarms of humans and robots will operate as a cohesive team with robots preventing humans from coming in harms way (Force Protection) and extending and amplifying their reach to allow 1 human to do the work of 10 humans (Force Multiplication). Our research will create swarms that will provide on-demand services in these missions.  The team is led by the University of Pennsylvania and includes collaborators from Penn and the U.S. Army Research Laboratory, the Massachusetts Institute of Technology, Georgia Institute of Technology, University of California and University of Southern California.

ARL CRA DCIST

Machine Learning and Control

Machine Learning and Control

Supported by: l4dc.org

Over the next decade, the biggest generator of data is expected to be IoT devices which sense and control the physical world. This explosion of data that is emerging from the physical world requires new ways for making sense of the data as well as making data-driven decisions. These challenges will require a rapprochement of areas such as machine learning and control theory,that have evolved independently over the past couple of decades. This project aims at connecting these two intellectually distant communities by an interdisciplinary approach that spans and connects the forefronts of robust control, deep learning, dynamical systems, nonlinear control theory, system identification, statistical learning, (deep) reinforcement learning, and convex optimization. A major thrust of our proposed research will create the foundation for statistical learning for dynamical and control system identification, approximation and abstraction, which will result in providing rigorous modeling guarantees using finite amount of data. Another major thrust focuses on rethinking nonlinear control analysis and stability from model driven to data driven, enabling machine learning to scale nonlinear control towards large scale, unknown models. Finally, the last major thrust focuses on rethinking robust control for providing robustness guarantees for deep learning components as well as in deep learning the context of feedback control loops. Recently, we also started a new conference on this exciting topic (l4dc.org)

l4dc.org