Han Zhou

PhD student

The University of Waterloo

About Me

Hi! I’m an incoming PhD student in compuer science at the University of Waterloo. I am fortunately advised by Yang Lu. I received my Master’s degree in Computer Science from the University of Hong Kong in 2020. Before that, I obtained my Bachelor’s degree in Info Management & Info System from Southwestern University of Finance and Economics in 2018. From 2020 to 2023, I worked as a data analyst at Baidu, ShenZhen.

My primary research focus is to enable scientific discoveries using trustworthy machine learning methods, specifically in the comprehension of biological data.

Interests

Computational Biology
Graph Neural Networks
Trustworthy Machine Learning
Natural Language Processing

Education

Master of Science in Computer Science, 2020

The University of Hong Kong
Bachelor of Management in Info Management & Info System (Financial Intelligence), 2018

Southwestern University of Finance and Economics

Publications

SONATA: Disambiguated manifold alignment of single-cell data

Rapid advancements in single-cell sequencing technologies have empowered biologists to collect high-resolution omics profiles at the cellular level. However, current methods recklessly report an arbitrary alignment solution without considering whether the data can be aligned. SONATA mitigates the ambiguity of manifold alignment by identifying potentially multiple valid solutions. (biorxiv)

Han Zhou, Kai Cao, Yang Young Lu

PDF Code

Connecting Embeddings Based on Multiplex Relational Graph Attention Networks for Knowledge Graph Entity Typing

Knowledge graph entity typing is a significant subtask of KG completion. However, rich relational knowledge is underutilized or even ignored in most existing work. We proposed a multiplex relational graph attention neural network (MRGAT), to integrate the heterogeneous structured information including entity typing tuples and entity relation triples.(TKDE 2022)

Yu Zhao, Han Zhou, Anxiang Zhang, Ruobing Xie, Qing Li, Fuzhen Zhuang

PDF

Entity Typing on Heterogeneous Relational Graph

Incorporating Global Information in Local Attention for Knowledge Representation Learning

Graph Attention Networks(GATs) lack the capability to model global importance relative to other entities of Knowledge Graphs, and we allow the proper incorporation of global information into the GAT family of models through the use of scaled entity importance, which is calculated by an attention-based global random walk algorithm. (Findings of ACL 2021)

Yu Zhao*, Han Zhou*, Fuzhen Zhuang, Qing Li, Ji Liu, (* equal contribution)

PDF Code

Recent News

Nov. 30, 2023 We present our work SONATA as a poster at MLCB 2023.