Zory Zhang

Email: [email protected]

Website: zoryzhang.notion.site

Nationality: China

Pronoun: he/him

Education

University of Illinois, Urbana-Champaign (UIUC) Aug 2023 - May 2025/6 (Expected)

• Pursuing Bachelor's degree of Science in Mathematics and Computer Science, CGA=3.96/4.00

Hong Kong University of Science and Technology (HKUST) Sep 2021 - July 2023

• Former undergraduate majoring in Computer Science and Mathematics

Research Interest

I’m interested in the computational modeling of human learning, which includes concept learning, language acquisition, learning to reason, social learning, causal learning, procedural learning, strategies to learn (e.g. via play), and even the development of these learning abilities.

Through reverse-engineer human learning, I attempt to join the modern attempt to answer the age-old question: how much innate knowledge and concepts is likely needed (e.g. does it includes language and which part if any?) to give rise to a certain amount of learning ability in certain aspects, and what is the right way (normatively and descriptively) to use them to help with learning. I have a board interest in borrowing insights from research in computational modeling and rational analysis, language acquisition, developmental and cognitive psychology, philosophy of cognitive science and mind, traditional and deep learning, and cognitive architecture, …

My computational modelling toolkit includes traditional and deep machine learning, hierarchical incremental concept formation (COBWEB), binding theory via synchrony + Hebbian learning (LISA), and explanation-based learning (EBL). I have worked on domains like learning to prove formal mathematical theorems, learning to answer complex logical queries on knowledge graphs, and learning concepts of patterns from knowledge graphs.

📄 Publications

• Dylan Zhang, Curt Tigges, Zory Zhang, Stella Biderman, Maxim Raginsky, and Talia Ringer (2024). Transformer-Based Models Are Not Yet Perfect At Learning to Emulate Structural Recursion. Transactions on Machine Learning Research. Arxiv.

Experience

Research Intern

May 2024 - Aug 2024. Institute for the Study of Learning and Expertise & Georgia Institute of Technology.

• Supervised by Pat Langley and Chris MacLellan.
• Work in progress: To computationally model human learning on structured domains (where information resides in not just entities but also relations), we’re developing a human-like (unsupervised, incremental, sample efficient, robust to noise) learning algorithm that can learn from patterns in knowledge graphs and gain the ability to inductively predict missing information in incomplete unseen knowledge graphs. This work follows along the line of “cognition as partly categorization” but increases the complexity of instances being categorized.
• My role: The only student project member. From problem formulation to literature review, to exploring design spaces, to careful experimentations, to report articulation. Presented progress report on COBWEB interest group meetings.

Research Project Leader, as a volunteer

Feb 2024 - Present. University of Illinois, Urbana-Champaign.

• Supervised by Jiaxuan You.