Gongjin Lan

About

Gongjin Lan is a versatile and prolific researcher whose work spans artificial intelligence, evolutionary robotics, computer vision, and federated learning. With a strong foundation in bio-inspired computation, Lan has made significant contributions to the field of modular robotics, particularly in developing algorithms that enable robots with evolvable morphologies to learn directed locomotion — a challenge that sits at the frontier of physically evolving robot systems. His 2021 work on Bayesian-Evolutionary optimization (65 citations) demonstrates his expertise in hybrid intelligent algorithms that balance exploration and efficiency in complex search spaces. Lan's research extends meaningfully into practical AI applications: his work on real-time vision for low-performance hardware and evolving compact deep neural networks addresses critical constraints in deploying intelligent systems on small robots. More recently, he has turned toward human-centered AI, with a highly cited 2024 review (66 citations) examining AI technologies in early childhood education, and contributions to personalized federated learning for facial expression recognition in care robotics. Collectively, his publications reflect a researcher committed to bridging theoretical machine learning with real-world robotic and educational applications, making him a noteworthy figure across multiple emerging AI disciplines.

Research Focus

Key Achievements

Top Papers

1. 1
   The key artificial intelligence technologies in early childhood education: a review

   66 citations · 2024
2. 2
   Time efficiency in optimization with a bayesian-Evolutionary algorithm

   65 citations · 2021
3. 3
   Learning directed locomotion in modular robots with evolvable morphologies

   26 citations · 2021
4. 4
   Directed Locomotion for Modular Robots with Evolvable Morphologies

   24 citations · 2018
5. 5
   Real-Time Robot Vision on Low-Performance Computing Hardware

   20 citations · 2018
6. 6
   Evolving Efficient Deep Neural Networks for Real-time Object Recognition

   20 citations · 2019
7. 7
   Learning locomotion skills in evolvable robots

   15 citations · 2021
8. 8
   Evolutionary predator-prey robot systems

   14 citations · 2019
9. 9
   Learning Directed Locomotion in Modular Robots with Evolvable Morphologies

   14 citations · 2020
10. 10
    Efficient lower layers parameter decoupling personalized federated learning method of facial expression recognition for home care robots

    11 citations · 2024

Key Collaborators

Contact & Links