The three main threads of DIL's research are (1) creativity in design of physical systems, (2) design of self-adaptive software agents, and (3) visual reasoning in cognition. The main research themes are (a) functional/teleological models of physical systems and software agents, (b) model-based reasoning & learning, (c) case-based and analogical reasoning & learning, and (d) classification and classification learning.
DIL has pioneered research on (1) structure-behavior-function (SBF) teleological models of physical systems and their use in analogical reasoning and learning in innovative design, (2) task-method-knowledge (TMK) teleological models of software agents and their use in reflection, introspection, and self-adaptation, and (3) visual analogies in problem solving, diagram understanding and design.
Current Projects
The current projects are listed here roughly in order of their "age" or "maturity."
Diagram Understanding and Knowledge Acquisition by Analogy: The Archytas Project. The goal is to develop an autonomous computer system that can understand design drawings, such as the drawings in a patent database, and acquire teleological models of designs by transferring and adapting the teleological model of a known drawing. Research issues include representation of drawings and their teleological models in an abstraction hierarchy, analogical mapping between the input drawing and the known drawings, and analogical transfer of the teleological model from a known drawing to the input drawing. Project members include Patrick Yaner and Michael Helms. Patrick has just completed his Ph.D. dissertation based on this work.
Meta-Reasoning in Classification Learning: The AN Project. The goal is to develop an autonomous software agent that automatically repairs its classification knowledge by making perceptually-verifiable predictions about the world, such as predictions about the stock market, checking those predictions over time, reflecting on its classification knowledge, and self-diagnosing and self-repairing the classification knowledge. Research issues include the representation and organization of the classification knowledge to support reflection and self-adaptation. Project members include
Joshua Jones and Ajay Chaudhari. Josh is working on his Ph.D. based on this project.
Story Construction in Intelligence Analysis: The STAB Project. The research goal is to develop an interactive intelligence assistant that can abduce a coherent story from large amounts of data, where the data is heterogenous, constantly evolving, and often conflicting. The research issues include representation and organization of stories, retrieval of story plots, and composition of new plots. We are also intergating the STAB system with an interactive information visualization tool called Jigsaw
developed by the Information Interfaces group. Project members include Summer Adams, Avik Sinharoy, Neha Sugandh and Anushree Venkatesh. This is a joint project with Prof. John Stasko,
Prof. Anita Raja (UNCC), and the Southeastern Regional Visual Analytics Center.
Learning Mental Models of Complex Systems: The ACT Project. The research issues are what are the mental models of complex systems that middle school children have, and how might we enable learning of deeper models of complex systems, such as an aquaria. Project members include Swaroop Vattam, Vivek Menon, and Brian Sherwell. This is a joint project with Dr. Spencer Rugaber,
Prof. Cindy Hmelo-Silver (Rutgers University) and Prof. Rebecca Jordan (Rutgers University).
Self-Adaptive Game-Playing Agents: The GAIA Project. The research issues are how might an autonomous game-playing software agent reflect on its decisions and actions (e.g., attacking an city in the interactive strategy game called FreeCiv) and repair its decision-making knowledge, and how might the agent assist game designers in making adaptations to designs of game-playing software agents. Project members include Joshua Jones, Chris Parnin, Praful Rana, Derek Richardson,
Avik Sinharoy and Jason Taylor. This is a joint project with Dr. Spencer Rugaber.
Biologically-Inspired Design Innovation: The DANTE Project. The research issues are how do teams of engineers and biologists generate new design ideas, and how might an interactive design assistant support the generation of new ideas by analogical retrieval and mapping of biological designs to engineering problems. Project members include Swaroop Vattam, Michael Helms, Vivek Menon and Iulian Radu. This is a joint project with the
Center for Biologically Inspired Design .
Cognitive Models of Autism. The research questions are what kind of information processing may explain autistic behavior, and how may we use computational tools for better communication with people on the autism spectrum. We are developing a cognitive model of autism to inform the design of computational tools for better commuication with autistic people. Project members include Maithilee Kunda.
Current Grants
Current research in DIL is supported by the following grants:
Learning about Causal Models of Complex Systems. PI: N. Hari Narayanan (Auburn); Co-PIs: Ashok Goel, Cindy Hmelo-Silver (Rutgers), Teresa Hubscher-Young (RPI) & Sadhana Puntambekar (Wisconsin). NSF SLC Catalyst Grant.
Multi-Modal Case-Based Reasoning in Modeling and Design. PI: Ashok Goel. NSF IIS
Grant.
Visual Analytics. PI: William Ribarsky (UNCC);
Georgia Tech PI: John Stasko, Georgia Tech Co-PIs: James Foley, Ashok Goel, Rebecca Grinter. DHS Grant under the auspices of NVAC (PNNL) and the Southeastern
Regional Visual Analytics Center .
Teleological Reasoning in Adaptive Software Design. PI: Ashok Goel; Co-PI: Spencer Rugaber.
NSF SoD Grant.
Learning about Complex Systems by Constructing Structure-Behavior-Function Models in Middle Schools. PI: Ashok Goel. Co-PI: Spencer Rugaber,
Cindy Hmelo-Silver (Rutgers), and Rebecca Jordan (Rutgers). NSF ALT Grant.
Towards a Computational Model of Biological Analogies in Innovative
Engineering Design. PI: Ashok Goel. NSF CreativeIT SGER Grant.