About me
Hi, there! I am Zhengyan Shi, a final-year undergraduate student majoring in Computer Science and Technology at Shanghai Jiao Tong University. I am working as a research assistant at SJTU-ReThinkLab, collaborating with Dr. Liangliang Shi, under the advisory of Prof. Junchi Yan. Moreover, I am working as a research intern at Institute of Artificial Intelligence (TeleAI), China Telecom, under the supervision of Dr. Dell Zhang.
Research
I am now interested in the intersection of machine learning and combinatorial optimization (ml4co) field, focusing on new solutions by a combination of diffusion framework and optimal transport. Previouly, I conducted research on self-supervised learning and multimodal learning, specifically on the application of optimal transport algorithms and generative models to graph matching, knowledge distillation and contrastive learning, etc.
Publications
SelKD: Selective Knowledge Distillation via Optimal Transport Perspective
Improving the Contrastive Learning via Graph Matching Perspective
Min: Mixture of Noise for Pre-Trained Model-Based Class-Incremental Learning
DB-OT: Optimal Transport with Double Bounds and its Applications in Clustering, Classification and Contrastive Learning
Education
Shanghai Jiao Tong University
B.Eng. in Computer Science and Technology
Chinese University of Hong Kong
Ph.D. in Computer Science and Engineering
To know me better
Hobbies
I am a huge fan of R&B and Cantopop, and Khalil Fong is my favorite singer. Songs of David Tao and Eason Chan also appeal to be a lot. I love singing (also good at, I think😋), because it makes me relaxed at any time. Also, sharing feelings and tastes about music with friends is enjoyable for me.