To design a better AI, we should still learn from brains.
- M.Sc., Electrical Engineering and Information Technology, Gottfried Wilhelm Leibniz Universität Hannover, Germany, 2019
- B.E., Mechatronics, Tongji University, China, 2016
Bldg 1, Level 4, 4335-WS08
Deyao Zhu is a Computer Science Ph.D. student at the Visual Computing Center (VCC) in King Abdullah University of Science and Technology (KAUST) under the supervision of Professor Mohamed Elhoseiny in the Vision-CAIR Group.
Education and Early Career
Deyao obtained his Master's degree in Electrical Engineering and Information Technology at Gottfried Wilhelm Leibniz Universität Hannover, Germany. His Master thesis is done in Max Planck Institute for Intelligent Systems which focuses on representation learning. Before that, he received his Bachelor’s degree in Mechatronics at Tongji University, China
He is interested in research that makes AI accumulate knowledge continuously to become smarter and smarter in reinforcement learning setups like video games, robotics, and autonomous driving.
- Value Memory Graph: A Graph-Structured World Model for Offline Reinforcement Learning link
- Social-Implicit: Rethinking Trajectory Prediction Evaluation and The Effectiveness of Implicit Maximum Likelihood Estimation link
- RelTransformer: A Transformer-Based Long-Tail Visual Relationship Recognition link
- Motion Forecasting with Unlikelihood Training in Continuous Space link
- Multimodal Trajectory Forecasting with Hallucinative Intents link