Health and Behavior: Evolving Adaptive Neural Systems
Led by Yufei Tang, Ph.D.
Yufei Tang is an Assistant Professor in the Department of CEECS and a Faculty Fellow of I-SENSE at Florida Atlantic University (FAU), where he is also the director of the Intelligent and Resilient Systems (IRS) Research Group. He received his Ph.D. in Electrical Engineering from the University of Rhode Island (URI) in 2016. His research interests are Machine Intelligence and Cyber-Physical Systems. In particular, he is interested in developing new intelligent algorithms (e.g., deep machine learning, large-scale network data mining) and building efficient and resilient cyber-physical Energy systems (e.g., renewable energy control and monitoring, critical infrastructure systems resilience).
The field of artificial intelligence has contributed great advances throughout nearly every aspect of society.
We are beginning to see an era where every person has their own personal assistant in their pocket, where vehicles will begin to take human error out of the roadways, and where emotional support in the form of an always on therapist is only a swipe away. It can be said that the ultimate goal within the field of AI, is to discover the general principles of learning so that an artificial agent may operate in this world of its own volition. The optimal path followed will necessarily involve the incorporation of that same mechanism by which our own form of intelligence came to be: evolution. Without evolution, it could be said that life would have remained as simple single-celled organisms in the microscopic scale without complex behaviors or the general learning mechanisms through which this proposal is written. Thus, given sufficient computational resources with sufficiently complex frameworks, the first artificial general intelligence is likely to arise out of an evolutionary search through the possible architectures of neural networks similar to our own.
As such, this project seeks to contribute to the development of more complex and complete computational models of evolution for artificial neural networks in order to widen the scope of tasks a given model can accomplish, reduce the problem of catastrophic forgetting, and discover the principles of general learning by which all intelligent life operates. To this end, the selected REU will be working with the team (i.e., a graduate student and the advisor) in the field of Evolved Plastic Artificial Neural Networks (EPANNs). This REU will be brain storming with the team and helping the graduate student preparing dataset, implementing algorithms, carrying our experiments, and writing technical paper. The expected REU should be strong in math, with coding experience a plus. The project will provide a meaningful experience for the participant, while contributing to Dr. Tang’s ongoing work in this area.