Mr. Haoyan Fu, Beijing Institute of Technology, China
Mr. Haoyan Fu is currently pursuing a Master of Artificial Intelligence at Beijing Institute of Technology, expecting to graduate in June 2025 with a GPA of 3.58/4. He earned his Bachelor’s degree in Software Engineering from the same institution in June 2022. His research experience includes developing a novel time-guided diffusion model for time-aware sequential recommendation and creating a dynamic hypergraph learning paradigm to improve session-based recommendation frameworks. His innovative work showcases his commitment to advancing artificial intelligence technologies.
Professional Profile:
Suitability for the Award:
Mr. Haoyan Fu stands out as an exceptional candidate for the Best Researcher Award due to his focus on cutting-edge methodologies in AI, specifically in the context of recommendation systems. His ability to innovate within the realm of Graph Neural Networks and Hypergraph Learning, combined with his strong academic background and published work in high-impact journals, makes him a worthy contender. The depth of his research aligns with industry needs for more accurate and scalable recommendation solutions, and his contributions to the field of AI-driven recommendations are likely to have long-term implications for both academia and industry.
Educational Background:
Mr. Haoyan Fu is currently pursuing a Master of Artificial Intelligence at Beijing Institute of Technology, with an expected graduation date in June 2025, maintaining a GPA of 3.58/4. He previously earned his Bachelor’s degree in Software Engineering from the same institution, graduating in June 2022 with a GPA of 3.56/4.
Technical Skills:
Haoyan is proficient in multiple programming languages, including Java, C++, Python, C, and SQL. He is skilled in utilizing advanced tools and frameworks such as TensorFlow, PyTorch, Scikit-Learn, NumPy, Pandas, Jupyter, and CUDA.
Research Experience:
- Time-Aware Sequential Recommendation (Mar 2024 – Aug 2024) ⏳
Haoyan proposed a novel time-guided diffusion model to address temporal sparsity in sequential data, developing a generalized Graph Neural ODE for aligning dynamic user and item representations over time.
- Dynamic Hypergraph Learning for Session-based Recommendation (Aug 2023 – Feb 2024) 🔄
He created a dynamic learning paradigm that transitioned from “graph” to “hypergraph,” integrating key relationships into session-based recommendation frameworks to enhance static graph information.
Publication Top Notes:
-
Title: Fusing Temporal and Semantic Dependencies for Session-Based Recommendation
-
Title: Diversified Graph Recommendation with Contrastive Learning
-
Title: Towards Relevance and Diversity in Crowdsourcing Worker Recruitment with Insufficient Information