Junjie Shentu

Postgraduate Student in the Department of Computer Science

What do you do?

I am a Ph.D. candidate in the Department of Computer Science. I come from China. My research interests are computer vision, deep generative models, and the application of generative models in various domains including medicine and engineering.

How are you involved in this area of science? 

Deep generative models have shown great merits and importance in artificial intelligence, offering tremendous help in various applications. Inspired by the success of deep generative models, I am making some algorithmic innovations to boost the model performance on specific tasks and the ability to generalize across different tasks. Moreover, I also managed to apply deep generative models in practical applications, such as the diagnosis of medical images and the augmentation of medical image databases.

What do you love about this topic?

There are several reasons for me to love this topic. First, as a data-driven approach, deep generative models offer a new solution to problems that were too hard and complicated to be addressed before. Second, this is a hot topic attracting a lot of researchers who are constantly making innovations, so there are numerous new lore to be learned. Last but not least, applying deep generative learning models in the practical solutions can offer a good comparison and complement to the traditional methods.

How does this work deliver real-world impact?

Deep generative models have presented significant effects in the real world. As a deep generative model for natural language processing, ChatGPT has already shown its great power in handling multiple tasks, such as question answering and document summarization. In terms of visual tasks, the deep generative model can produce high-quality images according to text description, boosting creativities and saving a lot of work for artists. In addition, its applications in other domains present tremendous assistance to human beings.