College of Science & Engineering
This research group studies a variety of problems in artificial intelligence and multi-agent systems. The problems can require machine learning methods, optimization algorithms, search algorithms that work in very large search spaces, image processing, disitributed computations, and more. When the search space is very large or the training datasets are huge, the computing resources have to scale up to the size of the problem, necessitating the use of MSI resources. Projects being undertaken include:
- Large scale simulation of mobile robots in a swarm. The simulator models physics in the interactions of the robots with the environment, so it takes significant computing cycles. The researchers will run at least 1,000 robots and possibly more to see how well the algorithms scale.
- Learning methods for a game theoretic approach where two agents play sequences of games against each other and try to learn a model of the opponent.
- Using deep learning methods to extract information from videos of the interactions of toddlers with a small humanoid robot. The interactions are part of a session designed to assess potential developmental disorders, such as autism spectrum disorder, of the child. The analysis of the videos aims at tracking the motions of the toddler to measure, over the interaction period, the distance of the child to the robot, the caregiver, and the robot experimenter. Social distances provide important information for behavior analysis. These tasks are computationally intensive and require storage space to store the videos that have to be processed and other data that are used for training.