Yolanda Gil (PI), Awards and Grants

• Designing Scientific Software One Workflow at a Time
  National Science Foundation (NSF).
  Grant number CCF-0725332. October 2007 - September 2010.
  Yolanda Gil (PI) and Ewa Deelman (co-PI).
  Much of science today relies on software to make new discoveries. This software embodies scientific analyses that are frequently composed of several application components and created collaboratively by different researchers. Computational workflows have recently emerged as a paradigm to manage these large-scale and large-scope scientific analyses. Workflows represent computations that are often executed in geographically distributed settings, their interdependencies, their requirements and their data products. The design of these workflows is at the core of today's scientific discovery processes and must be treated as scientific products in their own right. The focus of this research is to develop the foundations for a science of design of scientific processes embodied in the new artifact that is the computational workflow. The work will integrate best practices and lessons learned in existing workflow applications, and extend them in order to define and formalize design principles of computational workflows. This work will result in a fundamentally new approach to designing workflows that will greatly improve the scientific software design methodology by defining and formalizing design principles, and by familiarizing the scientific community with these effective workflow design processes.
• Plato: Phased-Learning through Analyzing Teaching and Observation
  DARPA Bootstrapped Learning (BL) program.
  Grant number HR0011-07-C-0060, subcontract to SRI International. August 2007-July 2011.
  ISI co-PIs: Paul Cohen and Yolanda Gil.
  The goal of this project is to develop an electronic student that can learn from a teacher using different methods of natural instruction. We will contribute the strategies to learn from being told by the teacher a broad range of generalities about process knowledge. These general descriptions will be tested by the learner with examples and practice of those processes. We will use Interdependency Models to relate the individual teacher instructions, check the consistency with the student's prior knowledge, and detect gaps in the stated instruction that could be filled through practice.
• Windward : Scalable Knowledge Discovery Through Grid Workflows
  Air Force Research Laboratory (AFRL).
  Grant number FA8750-06-C-0210. September 2006 - September 2010.
  Yolanda Gil (PI). ISI co-PIs: Paul Cohen and Ewa Deelman.
  Distributed workflows are emerging as a key technology to conduct large-scale and large-scope scientific applications in earthquake science, physics, astronomy, and many other sciences. In this new project, we will investigate the use of workflow technologies for Artificial Intelligence applications with a particular focus on data analysis and knowledge discovery tasks. Based on the data to be analyzed, an initial workflow template is formed by selecting from a library of known-to-work compositions of general-purpose machine learning algorithms. The workflow template is specialized through knowledge-based selection and configuration of algorithms. Finally, the workflow is mapped to available resources and restructured to improve execution time. Data analysis and knowledge discovery applications will benefit from the automation, scale, and distributed data and resource integration supported by distributed workflow systems. We will also conduct new research in important aspects of workflow systems. To what extend can we represent complex algorithms and their subtle differences so that they can be automatically selected and configured to satisfy the stated application requirements? Can we develop learning techniques that improve the performance of the workflow system by exploiting an episodic memory of prior workflow executions? What mechanisms will be needed to support autonomous and robust execution of concurrent workflows over continuously changing data?.
• NSF Workshop on Challenges of Scientific Workflows
  National Science Foundation (NSF).
  Grant number IIS-0629361. May 2006 - October 2007.
  Yolanda Gil (PI) and Ewa Deelman (co-PI).
  In recent years, workflows have emerged as a paradigm for conducting large-scale scientific analyses. The structure of a workflow specifies what analysis routines need to be executed, the data flow amongst them, and relevant execution details. Workflows provide a systematic way to capture scientific methodology and provide provenance information for their results. Robust and flexible workflow creation, mapping, and execution are largely open research problems. Under this project, Ewa Deelman and Yolanda Gil chaired an invitation-only workshop on "Challenges of Scientific Workflows" at the National Science Foundation. The aim of this workshop was to bring together IT researchers and practitioners working on a variety of aspects of workflow management as well as domain scientists that use workflows for day-to-day data analysis and simulation. The National Science Foundation expects a final report with recommendations to the community regarding the challenges of scientific workflows and their role in cyber infrastructure planning for 21st century science and engineering research and education.
• C4ML : Metareasoning for Integrated Learning
  DARPA Integrated Learning (IL) program.
  Grant number FA8650-06-C-7606, subcontract to BBN Technologies. May 2006 - July 2010.
  ISI co-PIs: Paul Cohen and Yolanda Gil.
  In this project we will develop a learning metareasoner to coordinate the activities of many learners in an integrated system that learns procedural knowledge from user demonstrations and past knowledge. A learning metareasoner is a problem solver that has explicit representations of its current learning state, learning goals, and has metareasoning methods to accomplish those goals. The learning metareasoner will assess its progress based on four criteria: capability, confidence, coverage, and competence (C4).
• Intelligent Optimization of Parallel and Distributed Applications
  National Science Foundation (NSF).
  Grant number CSR-0615412. August 2006 ­ September 2009.
  Principal Investigators: Mary Hall (PI), Kristina Lerman (co-PI), Ewa Deelman (co-PI), Aichiro Nakano (co-PI), Joel Saltz (co-PI). ISI co-PIs: Yolanda Gil.
  This project will develop a domain-specific programming system supporting Petascale application optimization of molecular dynamics simulation, in which applications will be viewed as workflows consisting of composable components to be mapped to a diversity of machine resources. The application components will be viewed as dynamically adaptive algorithms for which there exist a set of variants and parameters that can be chosen to develop an optimized implementation. A variant describes a distinct implementation of a code segment, perhaps even a different algorithm. A paramater is an unbound variable that affects application performance. By encoding an application in this way, we can capture a large set of possible application mappings with a very compact representation. Because the space of mappings is prohibitively large, the system captures and utilizes domain knowledge from the domain scientists and designers of the compiler, run-time and performance models to prune most of the possible implementation. Knowledge representation and machine learning techniques utilize this domain knowledge and past experience to navigate the search space efficiently. Incorporating cognitive search techniques and taking advantage of parallel resources, these alternative implementations are searched automatically by tools to find a high-quality implementation.
• MathTrust : Mathematical Analysis of Trust and Deception.
  Air Force Office of Scientific Research (AFOSR).
  Grant number FA9550-06-1-0031. December 2005 - November 2008.
  Yolanda Gil (PI).
  Information systems such as the Web often include open information sources that have very varying quality, and may be subject to deception. The information is often of unknown origins and there is often no prior history with many of the sources that may be used to assess their reputation. This project will investigate how to represent, learn, and characterize the reputation and reliability of sources in an information system that collects from users theit individual trust ratings and derives over time their collective consensus trust. This project will analyze the factors that affect trust in sources, study how to capture user feedback, and develop algorithms to derive source reputation. Based on this model of trust, a mathematical analysis of source trust and deception will relate formally a number of salient factors that influence trust in information systems.
• Intelligent Design and Optimization of Parallel and Distributed Applications
  National Science Foundation (NSF) Computer Science Research program.
  Grant number CNS-0509517. July 2005 - December 2006.
  Mary Hall (PI), Kristina Lerman (co-PI), Ewa Deelman (co-PI), Aichiro Nakano (co-PI), Joel Saltz (co-PI).
  This project will explore automatic mapping of applications to parallel systems consisting of tens of thousands of processors. This project includes expert domain scientists in molecular dynamics simulation and phylogenetics, who have been developing scalable algorithms that can handle large irregular data sets using high-end computing platforms. Many of these algorithms are hand-tuned and optimized for particular target architectures. The goal of the project is to develop expressive representations of optimization parameters, appropriate learning techniques for exploring the combinatoric optimization space, and automated mapping techniques for performance optimization.
• Towards Cognitive Grids : Knowledge-Rich Grid Services for Autonomous Workflow Refinement and Robust Execution.
  National Science Foundation (NSF) Shared Cyberinfrastructure program.
  Grant number SCI-0455361. December 2004 - November 2006.
  Ewa Deelman (PI), Yolanda Gil (co-PI).
  This research combines Artificial Intelligence and Distributed Computing techniques to create knowledge-rich workflow services that can support the execution of large-scale scientific workflows. The main foundation will be provided by expressive formal representations of the application workflow and of the execution environment. These representations will support resource selection that will enhance application performance, resource reservation based on anticipated workflow needs, workflow repair capabilities in case of failures or in case of new resources coming on line.
• CALO-KA : Interactive Acquisition of User Advice in a Cognitive Assistant that Learns and Organizes (CALO)
  DARPA Personalized Assistant that Learns (PAL) program.
  Grant number NBCHD030010, subcontract to SRI International. May 2003 - May 2008.
  USC/ISI co-PIs: Yolanda Gil, Jerry Hobbs, Craig Knoblock.
  This work is part of a very large integrated effort involving more than twenty universities and other research institutions throughout the US in order to develop personalized assistants that learn. The goal of our research is to develop novel techniques to assist users in specify new knowledge for an automated assistant that learns to improve its performance over time. This requires new research on acquiring advice through natural language interaction, operationalizing user advice into procedural and task knowledge, dialogue management to ask follow up questions about the implications and potential side effects of the user advice, generalizing user advice based on past experience, and expanding the set of terms known to the system when confronted with an unexpected situation. This work also involves designing a meta-reasoning architecture that includes a memory structure to index past experiences, reasoning about the implications of changes to the system's current knowledge, anticipating user requests and potential failures and opportunities, and self-motivated learning goals to prompt focused knowledge acquisition.
• Just-In-caSe just-in-Time Information Analysis
  Grant number N66001-03-C-8006. December 2002 - December 2005.
  PI: Yolanda Gil.
  The goal of this research is to develop a Web-based environment for information analysis that provides an emerging self-organization of knowledge through the use of natural language and machine learning techniques, including topic detection and similarity-based clustering. The system will exploit this emerging organization to support users to work in new topics, to debate alternative hypotheses, and to locate trustworthy open web sources.
• SCEC/IT: An Information Infrastructure for System-Level Earthquake Research.
  National Science Foundation (NSF), ITR Large Grant.
  Grant number EAR-0122464. September 2001 - September 2006.
  PI: Thomas Jordan. USC/ISI co-PIs: Carl Kesselman, Yolanda Gil, Hans Chalupsky. UCSD co-PI: Jean Bernard Minster. SDSC co-PI: Reagan Moore.
  This NSF Large ITR project is a collaboration with members of the Southern California Earthquake Center (SCEC) and funds a variety of information technologies to support earthquake research, including computational grid, digital libraries, ontologies and knowledge representation, planning, and interactive acquisition.
• TRELLIS: Capturing and Exploiting Semantic Relationships for Information and Knowledge Management.
  Air Force Office of Scientific Research (AFOSR).
  Award Number F49620-00-1-0337. August 2000 - November 2003.
  Yolanda Gil (PI).
  TRELLIS is an interactive environment that will allow users add their observations, opinions, and conclusions as they analyze information by making semantic annotations to documents and other on-line resources. This is in essence a knowledge acquisition problem, where the user is adding new knowledge to the system based on their expertise as they analyze information.
• TEMPLE: Template Enhancement through Knowledge Acquisition.
  DARPA Active Templates (AcT) program.
  Award Number F30602-00-2-0513. April 2000 - April 2003.
  Yolanda Gil (PI).
  The proposed work will develop an acquisition interface for planning knowledge that relies on script-based wizards to guide users in adding planning constraints and preferences.
• PHOSPHORUS: A Knowledge and Experience-Based Agent Capabilities Matcher.
  DARPA Control of Agent-Based Systems (CoABS) program.
  Award Number F30602-97-C-0068. June 1999 - December 2002.
  Yolanda Gil (co-PI) and Robert MacGregor (co-PI).
  We are developing Phosphorus, a knowledge-based matcher that accepts a user's description of a needed service as input and responds with a ranked list of agents that have the capability to provide that service. The Phosphorus matcher will exploit subsumption, goal reformulation, and partial match. The matcher will also be experienced-based, using learning techniques to improve the utility of its matches over time.
• KASPER: Knowledge Acquisition for Solving Problems.
  DARPA Rapid Knowledge Formation (RKF) program.
  Subcontract to SRI, Award Number N66001-00-C-8018. April 2000 - May 2003.
  Yolanda Gil (PI).
  This project will develop, in an integration effort with other research groups, tools to enable domain experts to extend knowledge bases by using natural language interfaces, commonsense reasoning, and analogy-based reasoning. Our group's contributions focus on tools to formulate follow-up questions when users have not provided sufficient knowledge, and on the acquisition of problem solving and process knowledge. The integrated system will be tested by two challenge problems designed by DARPA. One problem is based on how our system will acquire graduate level knowledge of biology from a textbook and then answer the questions at the end of the chapter. Another challenge problem will require developing expert-level knowledge-based techniques for advanced genome annotation and exploitation for pathogen countermeasures.
• KAMM: Knowledge Acquisition for Objective Grammars in MasterMind Editor.
  Air Force Research Laboratory's Joint Defense Planner (JDP) program.
  Award Number F30602-97-C-0118. April 2000 - September 2000.
  Yolanda Gil (co-PI) and Pedro Szekely (co-PI).
  This project is developing an editor that allows users to change and extend an initial grammar of objectives. This editor is integrated with the MasterMind objectives editor and includes acquisition wizards developed with EXPECT knowledge acquisition techniques. The editor was delivered in October 2000, and is now being integrated into the Global Command and Control System (GCCS) and is expected to be delivered to Air Operation Centers around the world by August 2001.
• SHERPA: Knowledge Acquisition for Large Knowledge Bases - Integrating Problem-Solving Methods and Ontologies into Applications.
  DARPA High Performance Knowledge Bases (HPKB) program.
  Award Number F30602-97-1-0195. April 1997 - December 2000.
  Bill Swartout (co-PI) and Yolanda Gil (co-PI).
  This work extends the ISI EXPECT architecture to include several novel approaches including the derivation and use of knowledge Interdependency Models, script-based knowledge acquisition, the integration of natural language techniques in knowledge acquisition tools, and the use of background knowledge to guide users in adding new knowledge to a system. We participated in the HPKB annual Challenge Problems, as well as in the Knowledge Acquisition Critical Component Experiment held at the Army Battle Command Battle Lab in Ft Leavenworth, KS, in August 1999. Several Army officers successfully used EXPECT's knowledge acquisition tools to extend the knowledge base.
• INSPECT-II: An Air Campaign Planning Evaluation Aid.
  DARPA Joint Forces Air Component Commander (JFACC) program.
  Award Number F30602-97-C-0118. April 1997 - September 2000.
  Yolanda Gil (co-PI) and Bill Swartout (co-PI).
  We extended the INSPECT air campaign plan critiquing tool that we had previously developed. Current funding is supporting thechnology transition to the Joint Defense Planner, which is on the path to becoming an integral part of TBMCS. INSPECT was originally developed under the DARPA Rome Laboratory Planning Initiative, and was demonstrated at the first US Air Force Expeditionary Force Experiment (EFX-98).
• ROSETTA: Ontology-Based Agent Communication.
  DARPA Information Systems Office (ISO) Technology Integration Experiment program.
  Award Number F30602-97-1-0195. July 1999 - July 2000.
  Yolanda Gil (co-PI) and Robert MacGregor (co-PI).
  The purpose of DARPA ISO Technology Integration Experiments is to investigate high-payoff links across DARPA ISO programs. Rosetta is a prototype message translation system that supports communication between heterogeneous agents using ontology merging technology and exploiting ontologies developed under the HPKB program to address inter-agent communication issues of central relevance to CoABS.
• EXPECT-II: A User-Centered Environment for the Development and Adaptation of Knowledge-Based Planning Aids.
  DARPA / Rome Planning Initiative (ARPI).
  Award Number DABT 63-95-C-0059. May 1995 - May 1999.
  Yolanda Gil (co-PI) and Bill Swartout (co-PI).
  EXPECT is a user-centered environment to develop and maintain knowledge bases. It includes knowledge acquisition tools, problem solving and reasoning modules, and a facility to generate natural language paraphrases of its knowledge. EXPECT was used to develop plan evaluation and critiquing systems for logistics planning and for air campaign planning. We participated in the Fourth Integrated Feasibility Demonstration (IFD-4) of the Planning Initiative, held at US Air Force Air Combat Command in June 1996 and in the Multi-Agent Planning, Visualization, and Simulation (MAPViS) integrated demonstration in 1998.