Haoran Zhu

I'm a PhD candidate in the Department of Electrical and Computer Engineering at New York University, advised by Prof. Anna Choromanska. I work on deep learning for autonomous driving.

My ultimate vision is to build intelligent machines for the physical world. To enable that long term vision, I'm currently focusing on autonomous driving because:

• autonomous driving is generally a relatively simpler problem, as it mainly involves navigation instead of harder navigation+manipulation for the general embodied AI tasks.
• level 2 autonomous driving is becoming increasingly popular in modern car systems, generating positive cash flow to support long-term level 4+ autonomous driving research.
• given the two reasons above, if we cannot solve autonomous driving, we are unlikely to solve the broader problem of embodied AI.

To unlock fully autonomous driving, my answer is building self-supervised world models for end-to-end learning.

Email  /  Scholar  /  Github  /  Linkedin

Research

I'm interested in deep learning for autonomous driving and general deep learning algorithms. In particular, I'm focusing on world models, self-supervised representation learning, end-to-end learning, multi-modal perception (camera, LiDAR, radar) and continual learning.

Miscellanea

Design and source code from Jon Barron's website