The Navy Center for Applied Research in Artificial Intelligence (NCARAI) hosts talks by leading Artificial Intelligence (AI) and Machine Learning experts from Academia, Private Sector, and the U.S. Government on a nearly weekly basis - and you can participate - for free!
The NCARAI “has been involved in both basic and applied research in artificial intelligence, cognitive science, autonomy, and human-centered computing since its inception in 1981. NCARAI, part of the Information Technology Division within the Naval Research Laboratory, is engaged in research and development efforts designed to address the application of artificial intelligence technology and techniques to critical Navy and national problems.” It’s four core capabilities are:
- Adaptive Systems,
- Intelligent Systems,
- Interactive Systems, and
- Perceptual Systems.
The next talk is scheduled for 5 March 2021 at 1200 Eastern / 0900 Pacific and Dr. Rosina Weber, Drexel University, will discuss Knowledge-based Explainable AI (XAI).
- Abstract: Underspecification in machine learning has been blamed for lack of accuracy. The idea is that there may be important aspects that connect features and instances that are not explicitly represented in data. Such limitation is obviously transferred to interpretable machine learning (IML) and explainable artificial intelligence (XAI) methods given they are based on attributing features or instances to explain decisions. Knowledge-based explanations seek to formulate the explanation process with domain knowledge to complement potentially underspecified data to produce explanations that make sense to users. In this talk, I demonstrate how to implement a knowledge based XAI approach in image, tabular, and textual data, discussing the potential benefits and challenges that make opportunities for further research.
- Presenter: Rosina Weber, Ph.D., Associate Professor in the Information Science Department at Drexel University, specializes in explainable AI (XAI), case-based reasoning (CBR), and textual AI agents. With over 100 peer-reviewed publications, one co-authored book, and three edited books, Dr. Weber has produced important contributions in use-inspired AI research in diverse fields such as legal, military, knowledge management, and scientometrics. Although having worked with various AI techniques, her main interest has focused on CBR projects including integrations and synergies between CBR and neural networks. Her research is currently funded by the NIH-NCATS Biomedical Data Translator to design and evaluate strategies to explain answers produced by an intelligent autonomous relay system where various agents execute multiple tasks in support of refining research hypotheses explored by translational scientists. The knowledge based XAI method proposed here was awarded best paper at the 2018 International Conference on CBR. She has served as program chair in numerous CBR and XAI events, having co-chaired the last AAAI Explainable Agency in AI workshop, and the two last IJCAI XAI workshops with NCARAI's director David W. Aha. She has also proposed the first and second mid-career diversity events at AAAI 2020 and 2021. She visits NRL for the first time since having left her position as a post doc with the NCARAI in 2001. She's been at Drexel University ever since.
To learn how to participate, go to the SLAMR Events calendars, click on “Knowledge-based Explainable AI (XAI)” on 5 March 2021 and then click “View Details.” Artificial Intelligence and Machine Learning community events are colored magenta.