The Full Story
The Clemson University INFORMS Student Chapter aims to promote interest in the fields of Operations Research, Management Science, and Analytics. We provide opportunities for academic, professional, and personal development for student members within the College of Business, the College of Science, and the College of Engineering, Computing and Applied Sciences.
Our mission is (i) to mentor students across campus interested in OR/MS; (ii) to build connections between students and professionals in both industry and academia; and (iii) to promote the exchange of ideas and to encourage collaboration between faculty and students from a range of disciplines, as well as with members of industry.
IE GRADUATE SEMINAR SERIES:
DR. MICHAEL DORNEICH (IOWA STATE UNIVERSITY)
CAN COMPUTERS ADAPT TO YOUR EMOTIONS? ADAPTIVE SYSTEMS IN INTELLIGENT TRAINING.
Fri, Nov 8 | 1:25 PM - 2:15 PM
Freeman Hall Auditorium
211 Fernow St, Clemson, SC 29634, USA
Adaptive systems are becoming more necessary as intelligent assistants are spreading into every aspect of work, education, and home life. Increasingly autonomous systems will have the authority and ability to initiate changes in the type of support they provide humans, based on its assessment of the current state of the human user. These systems will have a much more complete understanding of the user if it takes into account the user’s emotional state while they perform task. If a student becomes frustrated while trying to learn a task, a human tutor will adjust the way they talk to the student to help mitigate the emotional response. If a trainee becomes stressed while trying to learn a task, a human trainer will adjust the stress levels of the training to appropriate levels. In this work, we will discuss ways that intelligent tutoring and training systems can adjust their interactions with the human to account for frustration and stress. In one application, we investigated adapting the interaction style of intelligent tutoring system (ITS) feedback based on human–automation etiquette strategies. A rule set was developed that systemically selected the proper etiquette strategy to address one of four learning factors (motivation, confidence, satisfaction, and performance) under two different levels of user frustration. In a second application, we are developing an adaptive virtual reality (VR) stress training system that graduates the environmental stressors over time to inoculate stress and enhance performance. Physiological stress indicators are assessed to classify the moment-to-moment stress of the user and adapt the VR simulation to best train gradually to stressful spaceflight scenarios. If adaptive computer systems can vary their behavior to effectively mitigate negative emotions, then designers would have one more mechanism in which to design affect-aware adaptations that provide the proper responses in situations where human emotions affect the ability to learn.
DR. PHEBE VAYANOS (USC)
AI AND OPTIMIZATION
Mon, Nov 11 | 2:30 PM - 3:30 PM
Freeman Hall Auditorium
211 Fernow St., Clemson, SC 29634
With increase in the scale and diversity of deployment of AI- and optimization driven algorithms in the open world come several challenges including the need for tractability and resilience, issues of data scarcity and bias, information endogeneity, ethical considerations, and issues of shared responsibility between humans and algorithms. In this talk, we focus on the problems of homelessness, suicide prevention, public health, and disaster planning, and present research advances in AI and robust optimization to address one key cross-cutting question: how to effectively allocate scarce intervention resources in these domains while accounting for the challenges of open world deployment? We will show concrete improvements over the state of the art in these domains based on real world data and discuss steps we are taking to deploy these algorithms in the field to yield direct benefits for vulnerable communities.