About EMI Group

The Evolving Machine Intelligence (EMI) group was founded in September 2018 by Dr. Ran Cheng (PI), affiliated with Southern University of Science and Technology (SUSTech), China. Generally, our research motivations can be respectively interpreted in three levels:

  • Philosophically, we are motivated to understand how evolution generates complexity, diversity and intelligence;
  • Scientifically, we are motivated to study how intelligence can be made evolvable;
  • Technically, we are motivated to design learning/optimization computing paradigm across areas between evolutionary computation and deep learning, reinforcement learning, operation research, etc.

Our cutting-eadge research outcomes mainly  fall into the  interdisciplinary fields of evolutionary deep learning, evolutionary optimization, and evolutonary modeling, which  provide end-to-end solutions to applications in edge computing, intelligent manufacturing, etc.


Research Overview new 2
Evolutionary Deep Learning
Evolutonary Optimization
Evolutionary Modeling
Research Highlights
[ICCV 2021] Dense Video Captioning with Parallel Decoding
Teng Wang et al. Abstract: Dense video captioning aims to generate multiple associated captions with their temporal locations from the video. Previous methods follow a sophisticated ``localize-then-describe'' scheme, which heavily relies on numerous hand-crafted components. In this paper, we proposed a simple yet effective framework for end-to-end dense video captioning with parallel decod...
[ICCV 2021] FaPN: Feature-aligned Pyramid Network for Dense Image Prediction
Shihua Huang, Zhichao Lu, Ran Cheng*, Cheng He Abstract: Recent advancements in deep neural networks have made remarkable leap-forwards in dense image prediction. However, the issue of feature alignment remains as neglected by most existing approaches for simplicity. Direct pixel addition between upsampled and local features leads to feature maps with misaligned contexts that, in turn, tran...
[IEEE TCYB] Evolutionary Multiobjective Optimization Driven by Generative Adversarial Networks (GANs)
Cheng He, Shihua Huang, Ran Cheng*, Kay Chen Tan, Yaochu Jin Abstract: Recently, increasing works have been proposed to drive evolutionary algorithms using machine-learning models. Usually, the performance of such model-based evolutionary algorithms is highly dependent on the training qualities of the adopted models. Since it usually requires a certain amount of data (i.e., the can...