ai for intelligent training

Problem Formulation

Main problem-solving domains such as real-time damage control aboard naval vessels or emergency medical care require a human to make critical decisions in real-time with respect to several simultaneous phenomena. Such domains can be challenging due to the time and resource constraints and stressful due to high costs of making a mistake. Moreover, opportunities for real-life training are often limited as such training can be expensive and dangerous.

Our Contributions

Working in the ship-board damage control domain, we have developed a Petri-Net-based formalism as a rapid machine-learnable domain model whose predictions can be articulated to a trainee. We combined it with a machine-learned evaluator of domain states to predict and evaluate outcomes of a trainee's actions. Integrating these techniques into a blackboard-based problem-solver, we had a system with an expert-level performance and an ability to observe and critique trainee's actions.

On another project we combined an existing computational model for culturally affected behavior (CAB) with a subset of an appraisal-based emotion model (EMA). The resulting model, Culture, EMotion and Adaptation (CEMA), is a light-weight computational engine capable of adding culturally and emotionally affected behaviors to non-playable characters in immersive training systems, with the aim of increasing their believability and, consequentially, their training effectiveness.

Work in Progress

We are presently exploring the application of our player modeling techniques developed for the interactive storytelling and emotion modeling projects to intelligent training. First, we are looking at modeling the trainee's emotion state and using it to dynamically shape a medical training system for neonatal emergency care. Second, we are considering ways to model the student's emotional state and the learning style inclination in a massively open online course (MOOC).


  1. David Thue and Vadim Bulitko. 2018. Towards a Unified Understanding of Experience Management. In proceedings of the Fourteenth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE). Edmonton, Alberta. In press.
  2. Vadim Bulitko and Jessica Hong and Kumar Kumaran and Ivan Swedberg and William Thoang and Patrick von Hauff and Georg Schmölzer. 2015. RETAIN: a Neonatal Resuscitation Trainer Built in an Undergraduate Video-Game Class. arXiv:1507.00956 [cs.CY].
  3. Vadim Bulitko and Steven Solomon and Jonathan Gratch and Michael van Lent. 2008. Modeling Culturally and Emotionally Affected Behavior. In Proceedings of the Fourth Artificial Intelligence and Interactive Digital Entertainment Conference (AIIDE), pages 10-15. Stanford, CA.
  4. Vadim Bulitko and David C. Wilkins. 2006. ML-TIPN: An Algorithm for Automated Acquisition of Domain Models based on Time Interval Petri Nets. Journal of Multiple-Valued Logic and Soft Computing, volume 12, pages 391-407.
  5. Vadim Bulitko and David C. Wilkins. 2005. Machine Learning for Time Interval Petri Nets. Lecture Notes in Artificial Intelligence (LNAI), Proceedings of the Eighteenth Australian Joint Conference on Artificial Intelligence, pages 959 - 965. Springer-Verlag, Sydney, Australia.
  6. Vadim Bulitko and David C. Wilkins. 2003. Qualitative Simulation of Temporal Concurrent Processes Using Time Interval Petri Nets. Artificial Intelligence (AIJ), volume 144, issues 1-2, pages 95 - 124.
  7. Vadim Bulitko and David C. Wilkins. 1999. Automated Instructor Assistant for Ship Damage Control. In Proceedings of the Eleventh Innovative Applications of Artificial Intelligence Conference (IAAI), pages 778-785. Orlando, FL.
  8. Vadim Bulitko. 1998. Minerva-5: A Multifunctional Dynamic Expert System. M.Sc. thesis. University of Illinois at Urbana-Champaign. Pages 190.


  1. From Human Writers to AI Experience Managers. Liquid Narrative Group. North Carolina State University. Raleigh, North Carolina. October 8, 14.
  2. AI-based Interactive Experience Management. Invited keynote at Replaying Japan conference. August 23, 2014.
  3. AI-based Interactive Experience Management. GRAND workshop invited talk. University of Alberta. April 7, 2014.
  4. Experience Management with Artificial Intelligence. Lockheed Martin. October 18, 2013.
  5. Modeling Culturally and Emotionally Affected Behavior. The fourth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment. Stanford, California. October 22, 2008.
  6. Automated Instructor Assistant for Ship Damage Control. Institute for Creative Technologies (ICT), Marina del Rey, California. February 22, 2005.