UPDATE (2/24/25): Paper bidding has started
UPDATE (1/31/25): The date of the workshop is set at May 1, 2025, submission deadline updated to Feb. 21, 2025
UPATE (12/6/25): Andrea Stocco of UW will be this year’s keynote speaker!
UPDATE (12/6/25): Submission website now active.
Held at SDM-2025: https://www.siam.org/conferences-events/siam-conferences/sdm25/
Date: May 1, 2025 10:00am-3:30pm
Location: Alexandria, VA (The Westin Alexandria Old Town Hotel, room Edison C)
Part of SIAM Data Mining 2025 (SDM-25) (SDM occurs May 1-3, 2025 – full schedule here)
Thanks to our sponsor, SSCI for their support of this event.
Second Workshop on Metacognitive Prediction of AI Behavior
Organizing Committee:
Co-Chair: Paulo Shakarian, Arizona State University
Co-Chair: Nathaniel D. Bastian, Defense Advanced Research Projects Agency (DARPA) | United States Military Academy
Gerardo I. Simari, Universidad Nacional del Sur in Bahia Blanca, Argentina
Mario Leiva, Institute for Computer Science and Engineering, Argentina
Contact Information:
Paulo Shakarian, [email protected]; 699 S. Mill Ave., Tempe AZ 85281
Workshop Description:
As artificial intelligence (AI) becomes more prevalent in an increasing number of practical applications and systems, improved characterization of such systems will, in-turn, become important to ensure that such systems are resilient, safe, and reliable in the environments for which they are deployed – which may often differ from data used in training. However, while AI systems, often using supervised machine learning or reinforcement learning, have provided excellent results for a variety of applications, the reasons behind their failure modes – or anomalous behavior they engage in – are generally not well understood. The idea of metacognition, reasoning about an AI system itself, is a key avenue to understanding the behavior and performance of machine learning systems. Recently, a variety of methodologies have been explored in the literature, which include stress testing of robotic systems [1], model introspection [2], model certification [3], and performance prediction [4]. Moreover, researchers across multiple disciplines including computer science, control theory, mechanical engineering, human factors, and business schools have explored these problems from different angles. The objectives of the workshop are as follows:
- Survey various approaches to metacognition of AI systems
- Understand the requirements for various metacognitive approaches
- Identify novel methods for metacognition that drive improved AI performance in an operational or cross-domain setting
- Identify application areas suitable for the deployment of metacognitive methods
- Understand the relationship between AI metacognition and human operators
Specific topics to be covered include, but are not limited to:
- Explainable performance prediction of black-box AI systems
- Stress testing of reinforcement learning systems
- Test and Evaluation (T&E) of AI Systems
- How metacognition can be used to increase trust in AI systems by the operator
- Applications of AI metacognition to vision and robotic systems
- New methods leveraging neuro-symbolic AI architectures for metacognition
- Techniques for AI systems to self-adapt (self-heal, self-repair) in new domains
- Hyperdimensional Computing (HDC) and Vector Symbolic Architectures (VSA)
Call for Papers
We are currently accepting research papers in-line with the topics listed in the workshop description. Our goal is to accept high quality papers that include summaries previously submitted or published results relating to metacognition which may not be exposed to this new community (we just ask that you please cite the original version in your submission so readers can easily find it).
Papers will be non-anonymized and will be 2-4 pages in length using the format of IEEE Intelligent Systems. Papers not in the IEEE IS format may still be accepted provided they are of high quality as deemed by the reviewers. Please note reviewers will not be required to review content beyond 4 pages. Papers will undergo peer-review and we will aim to get all papers 2-4 reviews each.
Accepted papers will be posted to the workshop website, but this will not be considered archival.
Authors of selected paper will be invited to submit an extended version for a journal special issue or edited volume, details are forthcoming.
Click here for the paper submission website.
At least one author of each accepted paper must present in-person (and pay the necessary SDM registration).
Keynote
We are happy to announce Andrea Stocco, Associate Professor of Psychology at the University of Washingon as this year’s keynote speaker.
Key Dates
Feb. 21, 2025, AoE: Paper submission deadline (updated)
March 24, 2025: Notification of acceptance or rejection
May 1, 2025: METACOG-25 Workshop at SDM 2025
TBD: Submission deadline for extended journal version (this will occur several months after the event)
Workshop Format:
This workshop will run half-day.
Target Audience:
Based on the participation in our First Workshop on Metacognitive Prediction of AI Behavior (held at Arizona State in November 2023, sponsored by ARO), we anticipate broad, multi-disciplinary participation. During the first edition of this workshop, we had researchers representing computer science, cognitive psychology, electrical engineering, statistics, aerospace, and mechanical engineering, in addition to industry participation.
Summary of the Previous Workshop:
The first edition of the workshop was sponsored by the Army Research Office and took place Nov. 13-15, 2023 in Scottsdale, Arizona. Despite not being affiliated with a major conference event, there were still 25 presenters representing 14 universities and one commercial company. An edited volume created based on papers from the event is currently in-press with Cambridge University Press, and many of the videos have been posted to YouTube, receiving thousands of views. In addition to the presenters, representatives from DARPA, Armywh Research Office, and Raytheon were also in attendance. You can view the page for METACOG-23 here which includes videos from most of the talks at the event.
METACOG-25 Video
Biographies for Organizing Committee:
Paulo Shakarian, Ph.D. is a tenured Associate Professor in the School of Computing and AI at the Fulton Schools of Engineering at Arizona State University. He also holds the additional position as Research Director for the School of Computing and AI at ASU. He specializes in the fusion of symbolic artificial intelligence and machine learning – publishing numerous scientific books and papers. Shakarian was named a “KDD Rising Star,” received the Air Force Young Investigator award, received multiple “best paper” awards and has been featured in major news media outlets such as CNN and The Economist. Paulo has been funded by various organizations including IARPA (HAYSTAC, CAUSE, ICARUS), ARO (4x), ONR (5x), AF/AFOSR (2x), and DARPA as well as various industry partners. Paulo also co-founded a startup company that used machine learning to predict future exploits; the company was acquired after raising $8 million in venture capital and having obtained over 80 customers. Paulo also founded and manages the Neuro Symbolic Channel on YouTube which has over 2,500 subscribers. Earlier in his career, Paulo was an officer in the U.S. Army where he served two combat tours in Iraq, earning a Bronze Star and the Army Commendation Medal for Valor. During his military career, Paulo also served as a DARPA Fellow and as an advisor to IARPA. He holds a Ph.D. and M.S. in computer science from the University of Maryland, College Park, and a B.S. in computer science from West Point.
Nathaniel D. Bastian, Ph.D. is a Lieutenant Colonel in the U.S. Army, where he is an Academy Professor and Cyber Warfare officer at the United States Military Academy (USMA) at West Point, as well as a Program Manager at the Defense Advanced Research Projects Agency (DARPA). At USMA, Nate serves as Division Chief, Data & Decision Sciences and Senior Research Scientist at the Army Cyber Institute (ACI), as well as a faculty member with appointments in the Departments of Systems Engineering, Mathematical Sciences, and Electrical Engineering and Computer Science. At DARPA, Nate serves in the Information Innovation Office (I2O) developing, executing, and transitioning research and development programs in artificial intelligence and cyber operations. Nate has co-authored 100+ refereed journal articles, conference papers, book chapters, and textbooks. He is the recipient of numerous academic awards and honors, to include a Fulbright Scholarship and National Science Foundation Graduate Research Fellowship, and he has received $5M+ in research grants from multiple government organizations. Nate holds a Ph.D. in industrial engineering and operations research from the Pennsylvania State University, a M.Eng. in industrial engineering from PSU, a M.S. in econometrics and operations research from Maastricht University, and a B.S. in engineering management (electrical engineering) from USMA. His research interests aim to develop innovative, assured, intelligent, human-aware, data-centric, and decision-driven capabilities for cyberspace operations, command and control, and
Gerardo I. Simari, Ph.D., is a professor at Universidad Nacional del Sur in Bahía Blanca, and a researcher at CONICET, Argentina. His research focuses on topics within AI and Databases, and reasoning under uncertainty. He received a PhD in computer science from University of Maryland College Park and later secured a senior researcher position in the Department of Computer Science, University of Oxford (UK). His work was highlighted in the IJCAI 2019 Early Career Spotlight track, was selected as one of IEEE Intelligent Systems “AI’s Ten to Watch” for 2016, and received best paper awards at international scientific events. Simari is also a former Fulford Junior Research Fellow of Somerville College, University of Oxford.
Mario Alejandro Leiva is a researcher and educator with a strong background in computer science and data analysis. Holding a Doctorate in Computer Science from Universidad Nacional del Sur, he has contributed to the field of AI and computational reasoning. Currently serving as a Postdoctoral researcher at Instituto de Ciencias e Ingeniería de la Computación (ICIC), Universidad Nacional del Sur, CONICET, Leiva’s research primarily revolves around defeasible logic with applications to cyber threat analysis. Furthermore, Leiva has actively participated in various research projects. Alongside his research endeavors, Leiva is deeply engaged in academic instruction, having taught courses ranging from Algorithmics to Data Science.
References:
[1] H. Delecki, M. Itkina, B. Lange, R. Senanayake, and M. J. Kochenderfer, “How Do We Fail? Stress Testing Perception in Autonomous Vehicles.” arXiv, Mar. 26, 2022. Accessed: Feb. 28, 2023. [Online]. Available: http://arxiv.org/abs/2203.14155
[2] S. Daftry, S. Zeng, J. A. Bagnell, and M. Hebert, “Introspective Perception: Learning to Predict Failures in Vision Systems.” arXiv, Jul. 28, 2016. doi: 10.48550/arXiv.1607.08665.
[3] L. Li, T. Xie, and B. Li, “SoK: Certified Robustness for Deep Neural Networks,” presented at the 2023 IEEE Symposium on Security and Privacy (SP), Oct. 2022, pp. 94–115. doi: 10.1109/SP46215.2023.00006.
[4] P. Shakarian, A. Koyyalamudi, N. Ngu, and L. Mareedu, “An Independent Evaluation of ChatGPT on Mathematical Word Problems (MWP),” AAAI Spring Symposium, 2023.