E-mail and search functions

I-STEM Education Initiative

Return to I-STEM home page

Main Navigation

For those using screen readers: Disregard the following Javascript. It contains no content.

1. Campus P–16 STEM Education Outreach

CADENS: The Centrality of Advanced Digitally ENabled Science

National Science Foundation Award #1445176
Donna Cox
Art & Design
Dates: October 1, 2014–September 30, 2017 (estimated)

Computational data science is at a turning point in its history. Never before has there been such a challenge to meet the growing demands of digital computing, to fund infrastructure and attract diverse, trained personnel to the field. The methods and technologies that define this evolving field are central to modern science. In fact, advanced methods of computational and data-enabled discovery have become so pervasive that they are referred to as paradigm shifts in the conduct of science. A goal of this Project is to increase digital science literacy and raise awareness about the Centrality of Advanced Digitally ENabled Science (CADENS) in the discovery process. Digitally enabled scientific investigations often result in a treasure trove of data used for analysis. This project leverages these valuable resources to generate insightful visualizations that provide the core of a series of science education outreach programs targeted to the broad public, educational and professional communities. From the deep well of discoveries generated at the frontiers of advanced digitally enabled scientific investigation, this project will produce and disseminate a body of data visualizations and scalable media products that demonstrate advanced scientific methods. In the process, these outreach programs will give audiences a whole new look at the world around them. The project calls for the production and evaluation of two principal initiatives. The first initiative, HR (high-resolution) Science, centers on the production and distribution of three ultra-high-resolution digital films to be premiered at giant screen full-dome theaters; these programs will be scaled for wide distribution to smaller theaters and include supplemental educator guides. The second initiative, Virtual Universe, includes a series of nine high-definition (HD) documentary programs. Both initiatives will produce and feature data visualizations and the CADENS narratives to support an integrated set of digital media products. The packaged outreach programs will be promoted and made available to millions through established global distribution channels. Expanding access to data visualization is an essential component of the Project. Through a call for participation (CFP), the Project provides new opportunities for researchers to work with the project team and technical staff for the purpose of creating and broadly distributing large-scale data visualizations in various formats and resolutions. The project will feature these compelling, informative visualizations in the outreach programs described above. A Science Advisory Committee will participate in the CFP science selections and advise the Project team. The project calls for an independent Program Evaluation and Assessment Plan (PEAP) to iteratively review visualizations and the outreach programs that will target broad, diverse audiences.

The project launches an expansive outreach effort to increase digital science literacy and to convey forefront scientific research while expanding researchers access to data visualization. The project leverages and integrates disparate visualization efforts to create a new optimized large-scale workflow for high-resolution museum displays and broad public venues. The PEAP evaluations will measure progress toward project goals and will reveal new information about visualization's effectiveness to move a field forward and to develop effective outreach models. The project specifically targets broad audiences in places where they seek high-quality encounters with science: at museums, universities, K-16 schools, and the web. This distribution effort includes creating and widely disseminating the project outreach programs and supplemental educator guides. The project visualizations, program components, HD documentaries, educational and evaluation materials will be promoted, distributed and made freely available for academic, educational and promotional use. Dissemination strategies include proactively distributing to rural portable theaters, 4K television, professional associations, educators, decision-makers, and conferences. To help address the critical challenge of attracting women and underrepresented minorities to STEM fields, the Project will support a Broadening Participation in Visualization workshop and will leverage successful XSEDE/Blue Waters mechanisms to recruit under-represented faculty and students at minority-serving and majority-serving institutions and to disseminate the Project programs and materials among diverse institutions and communities.

Education Component: This project will also train graduate students in NLP and develop materials that can be used to teach middle and high school students about NLP and to inspire them to pursue an education in computer science.

CAREER: Bayesian Models for Lexicalized Grammars

National Science Foundation Award #1053856
Julia Hockenmaier
Computer Science
Dates: February 1, 2011–January 31, 2018 (estimated)

Natural language processing (NLP) is a key technology for the digital age. At the core of most NLP systems is a parser, a program which identifies the grammatical structure of sentences. Parsing is an essential prerequisite for language understanding. But despite significant progress in recent decades, accurate wide-coverage parsing for any genre or language remains an unsolved problem. This project will advance the state of art in NLP technology through the development of more accurate statistical parsing models.

Since language is highly ambiguous, parsers require a statistical model which assigns the highest probability to the correct structure of each sentence. The accuracy of current parsers is limited by the amount of available training data on which their models can be trained, and by the amount of information the models take into account. This project aims to advance parsing by developing novel methods of indirect supervision to overcome the lack of labeled training data, as well as new kinds of models which incorporate information about the prior linguistic context in which sentences appear. It employs Bayesian techniques, which give robust estimates and allow rich parametrization, and applies them to lexicalized grammars, which provide a compact representation of the syntactic properties of a language.

Education Component: This project will also train graduate students in NLP and develop materials that can be used to teach middle and high school students about NLP and to inspire them to pursue an education in computer science.

CAREER: Large-Scale Recognition Using Shared Structures, Flexible Learning, and Efficient Search

National Science Foundation Award #1053768
Derek Hoiem
Computer Science
Dates: May 1, 2011–April 30, 2017 (estimated)

This research investigates shared representations, flexible learning techniques, and efficient multi-category inference methods that are suitable for large-scale visual recognition. The goal is to produce visual systems that can accurately describe a wide range of objects with varying precision, rather than being limited to identifying objects within a few pre-defined categories. The main approach is to design object representations that enable new objects to be understood in terms of existing ones, which enables learning with fewer examples and faster and more robust recognition.

The research has three main components: (1) Designing appearance and spatial models for objects that are shared across basic categories; (2) Investigating algorithms to learn from a mixture of detailed and loose annotations and from human feedback; and (3) Designing efficient search algorithms that take advantage of shared representations.

The research provides more detailed, flexible, and accurate recognition algorithms that are suitable for high-impact applications, such as vehicle safety, security, assistance to the blind, household robotics, and multimedia search and organization. For example, if a vehicle encounters a cow in the road, the vision system would localize the cow and its head and legs and report "four-legged animal, walking left," even if it has not seen cows during training.

Education Component/Dissemination: The research also provides a unique opportunity to involve undergraduates in research, promote interdisciplinary learning and collaboration, and engage in outreach. Research ideas and results are disseminated through scientific publications, released code and datasets, public talks, and demonstrations for high school students.

Engineering Research Center for Power Optimization for Electro-Thermal Systems (POETS)

National Science Foundation Award #1449548
Andrew Alleyne
Mechanical Science and Engineering
Dates: August 1, 2015–July 31, 2020 (estimated)

Nearly all modern electronic systems are hitting a power density wall where further improvements in power density pose significant challenges. The NSF Engineering Research Center for Power Optimization for Electro-Thermal Systems (POETS), aims to enhance or increase the electric power density available in tightly constrained mobile environments by changing the design. The management of high-density electrical and thermal power flows is a safety-critical societal need as recent electrical vehicles and aircraft battery fires illustrate. Engineering education conducted in silos limits systems-level approaches to design and operation. POETS will create the human capital that is explicitly trained to think, communicate, and innovate across the boundaries of technical disciplines. The Engineering Research Center (ERC) will institute curricular reform to train across disciplines using a systems perspective. It will develop pedagogical tools that allow greater stems-level understanding and disseminate these throughout the undergraduate curriculum. POETS will target undergraduate curriculum modifications aimed at early retention and couple it with undergraduate research and K-12 teacher activities. POETS' research will directly benefit its industry stakeholders comprised of power electronics Original Equipment Manufacturers (OEM) and Small to Medium sized businesses in the OEM supply chain. An Industry/Practitioner Advisory Board will help direct efforts towards ready recipients of POETS research developments. POETS will harness the outputs of the ecosystem and drive research across the "valley of death" into commercialization.

POETS uses system level analysis tools to identify barriers to increased power density. Design tools will be used to create optimal system-level and subsystem-level designs. Novel algorithm tools will address the multi-physics nature of the integrated electro-thermal problem via structural optimization. Once barriers are identified, POETS will cultivate enabling technologies to overcome them. The operation of these systems necessitates development of heterogeneous decision tools that exploit multiple time scale hierarchies and are not suitable for real-time use. Implementation of these management approaches requires new 3D power electronics architectures that surpass current 2D designs. The thermal management will be tightly coupled with new 3D electronic systems designs using topology optimization for power electronics, storage, etc. The new designs will tightly interweave elements such as solid state thermal switches and modular multi-length scale elements; i.e. spreaders, storage units, phase change and mass flow system interacting with convection units. Fundamental research advances will support development of the 3D component technologies. New materials systems will be developed by manipulating nanostructures to provide tunable directionality for in plane and out-of-plane thermal power flows. These will be coupled with micro- and nano-scale thermal routing based on new conduction/convection systems. Buffers made from phase change material will be integrated into these systems to augment classes of autonomic materials with directed power flow actuation. Novel tested systems will integrate the system knowledge enabling technologies and fundamental breakthrough into modular demonstrations.

G.A.M.E.S. participants working in lab.
Girls Adventures in Mathematics, Engineering, and Science (G.A.M.E.S.)

Abbott Laboratories
Caterpillar Foundation
John Deere Foundation
Motorola Foundation Innovation Generation Grants
Shell Oil Company
Women in Engineering

University of Illinois / Urbana, Illinois – Girls Adventures in Mathematics, Engineering, and Science (G.A.M.E.S) Summer Camp is an annual week-long residential camp designed to give academically talented middle school girls an opportunity to explore math, science, and engineering careers through demonstrations, classroom presentations, hands-on activities, and contact with women in these technical fields.


Sustained-Petascale In Action: Blue Waters Enabling Transformative Science And Engineering

National Science Foundation Award #1238993
William Kramer
Computer Science
Dates: October 1, 2013–July 31, 2019 (estimated)

This a renewal award to the National Center for Supercomputing Applications (NCSA) at the University of Illinois at Urbana-Champaign (UIUC) to operate Blue Waters, which is a leadership class compute, network, and storage system, that will deliver unprecedented large scale and highly usable computing capabilities to the national research community. Blue Waters provides the capability for researchers to tackle much larger and more complex research challenges across a wide spectrum of domain than can be done now, and opens up entirely new possibilities and frontiers in science and engineering. This system is located at the newly constructed National Petascale Computing Facility at UIUC.

This award enables investigators across the country to conduct innovative research in a number of areas including: using three-dimensional, compressible, finite difference, magnetohydrodynamic (MHD) codes to understand how internal solar magnetoconvection powers the Sun's activity, and how that activity heats the chromosphere and corona and accelerates charged particles to relativistic energies; applying adaptive mesh refinement (AMR) technologies to study flows of partially ionized plasma in the outer heliosphere; implementing multiscale methods to study protein induced membrane remodeling key steps of the HIV viral replication cycle and clathrin coated pit formation in endocytosis; testing of the hypothesis that transport fluxes and other effects associated with cloud processes and ocean mesoscale eddy mixing are significantly different from the theoretically derived averages embodied in the parameterizations used in current-generation climate models; and, exploring systems-of-systems engineering design challenges to discover optimal many-objective satellite constellation design tradeoffs that include Earth science applications. Large allocations of resources on the new system have been awarded to scientists and engineers by NSF through a separate peer-reviewed competition.

The Blue Waters system and project are aligned with NSF's Advanced Computing Infrastructure Strategy to promote next generation computational and data intensive applications. These applications are being developed by multiple teams of researchers who will revolutionize and transform our knowledge of science and engineering across many disciplines. The system supports new modalities of computation, new programming models, enhanced system software, accelerator technologies and novel storage. The robust design and configuration of Blue Waters ensures that it will meet the evolving needs of the diverse science and engineering communities over the full lifetime of the system.

The broader impacts of this award include: provisioning unique infrastructure for research and education; accelerating education and training in the use of advanced computational science; training new users on how to use petascale computing techniques; promoting an exchange of information between academia and industry about the petascale applications; and broadening participation and collaborations in computational science with other research institutions and projects nationally and internationally.