Human brains are very complex and yet they are able to generalize in every domain with few samples. Current deep learning models, however, cannot generalize well for out of domain data. My research aims at this gap between human brains and current deep learning models and seeks to develop explainable models instead of black boxes. My focus is on developing new algorithms within the field of Deep Learning that can learn causal variables and their relations. I often work in collaboration with researchers in applied fields such as Autonomous Driving.
In the field of Intelligent Transportations Systems and Computer Vision, I have worked on learning predictive models for accident detection in roads and pedestrian detection in autonomous vehicles. I have also worked on blind spot detection in vehicles in the field of Data Fusion.
Biography
Shayan Shirahmad Gale Bagi is a PhD student supervised by Mark Crowley and Krzysztof Czarnecki in the Department of Electrical and Computer Engineering at the University of Waterloo and a member of the Waterloo Institute for Artifical Intelligence (waterloo.ai). He received his M.Sc. and B.Sc. in Electrical Engineering from the University of Tehran working on the advanced driver assistance systems (ADAS) in vehicles and pedestrian detection in autonomous vehicles (AVs).
Research Interests
- Deep Learning
- Autonomous Driving
- Data Fusion
- Causal Inference
- Machine Vision
