Also In This Section
  • Categories

  • Recent News

  • Day: August 5, 2020

    Tim Havens, Tony Pinar Co-Authors of Article in IEEE Trans. Fuzzy Systems

    An article by Anthony Pinar (DataS/ECE) and Timothy Havens (DataS/CC), in collaboration with University of Missouri researchers Muhammad Islam, Derek Anderson, Grant Scott, and Jim Keller, all of University of Missouri, has been published in the July 2020 issue of the journal IEEE Transactions on Fuzzy Systems.

    The article is titled, “Enabling explainable fusion in deep learning with fuzzy integral neural networks.” Link to the article here.

    Tony Pinar

    Tim Havens

    Information fusion is an essential part of numerous engineering systems and biological functions, e.g., human cognition. Fusion occurs at many levels, ranging from the low-level combination of signals to the high-level aggregation of heterogeneous decision-making processes. While the last decade has witnessed an explosion of research in deep learning, fusion in neural networks has not observed the same revolution. Specifically, most neural fusion approaches are ad hoc, are not understood, are distributed versus localized, and/or explainability is low (if present at all). Herein, we prove that the fuzzy Choquet integral (ChI), a powerful nonlinear aggregation function, can be represented as a multilayer network, referred to hereafter as ChIMP.

    We also put forth an improved ChIMP (iChIMP) that leads to a stochastic-gradient-descent-based optimization in light of the exponential number of ChI inequality constraints. An additional benefit of ChIMP/iChIMP is that it enables explainable artificial intelligence (XAI). Synthetic validation experiments are provided, and iChIMP is applied to the fusion of a set of heterogeneous architecture deep models in remote sensing. We show an improvement in model accuracy, and our previously established XAI indices shed light on the quality of our data, model, and its decisions.

    M. Islam, D. T. Anderson, A. J. Pinar, T. C. Havens, G. Scott and J. M. Keller, “Enabling Explainable Fusion in Deep Learning With Fuzzy Integral Neural Networks,” in IEEE Transactions on Fuzzy Systems, vol. 28, no. 7, pp. 1291-1300, July 2020, doi: 10.1109/TFUZZ.2019.2917124.

    Kelly Steelman Presents at ASEE

    Kelly Steelman, interim department chair and associate professor, Cognitive and Learning Sciences, presented her paper, “Work in Progress: Student Perception of Computer Programming Within Engineering Education: An Investigation of Attitudes, Beliefs, and Behaviors” at the 2020 ASEE Virtual Conference, June 22-26, 2020.

    Co-authors of the paper are Michelle Jarvie-Eggart (EF), Kay Tislar (CLS), Charles Wallace (CC), Nathan Naser (GMES), Briana Bettin (CS) and Leo Ureel (CS), all from Michigan Tech.

    Although most engineering faculty and professionals view computer programming as an essential part of an undergraduate engineering curriculum, engineering students do not always share this viewpoint. In fact, engineering students—especially those outside of computer and electrical engineering—may not realize the value of computer programming skills until after they have graduated and advanced in their career (Sterian, Dunne, & Blauch, 2005). Failure to find value in computer programming may have negative consequences for learning. Indeed, engineering students who do not view programming as interesting or useful show poorer performance on tests of programming concepts than students who do (Lingar, Williams, and McCord, 2017). This finding is consistent with theories of technology acceptance (e.g., Davis, 1989, Venkatesh, et al., 2003) that emphasize perceived usefulness as a key determinant of attitudes toward a technology and subsequent use or disuse of it. Accordingly, to better support student learning, engineering coursework should include specific interventions that emphasize the utility of programming skills for a career in engineering. Intervention effectiveness, however, may depend in part on the characteristics of the individual learners, including their prior programming experience, their openness to new experiences, and their beliefs about the nature of intelligence. The purpose of the current work is to understand engineering students’ attitudes toward and experiences with computer programming as well as to assess the relationship between their attitudes and experiences and their mindset toward their own intelligence. 101 engineering students participated in the study as part of a general education psychology course. Participants completed a computer language inventory and three surveys. The first survey inquired about students’ computer programming experiences and attitudes (Hoegh and Moskal, 2009). The second survey posed questions related to different aspects of openness to experience (Woo et al., 2014): intellectual efficiency, ingenuity, curiosity, aesthetics, tolerance, and depth. Finally, the third survey probed participants’ beliefs about the nature of intelligence and whether it is fixed or can be developed (Dweck, 1999). This paper will present the results of these surveys and explore the correlations among the various scales. The implications for engineering education interventions will be discussed.

    Download the paper here.

    Steelman, K. S., & Jarvie-Eggart, M. E., & Tislar, K. L., & Wallace, C., & Manser, N. D., & Bettin, B. C., & Ureel, L. C. (2020, June), Work in Progress: Student Perception of Computer Programming within Engineering Education: An Investigation of Attitudes, Beliefs, and Behaviors Paper presented at 2020 ASEE Virtual Annual Conference Content Access, Virtual On line .