Tiansi Dong

Dr. Tiansi Dong is a researcher in the Department of Computer Science and Technology at the University of Cambridge, leading innovative work on machine reasoning through geometric neural approaches. His research bridges symbolic AI and neural networks using sphere-based representations for explainable reasoning.

Core Research: Develops Sphere Neural Networks (SphNNs) that transform logical reasoning into geometric configurations, enabling rigorous neural-symbolic integration. Current projects focus on syllogistic reasoning, spatial cognition, and humor understanding as high-level cognitive tasks.

Published foundational works including Sphere Neural-Networks for Rational Reasoning (2024) and the monograph A Geometric Approach to the Unification of Symbolic Structures and Neural Networks (Springer, 2021). Organizes interdisciplinary workshops including NeurMAD@AAAI'25.

Teaching: Lectures on Explainable AI and neural-symbolic unification at Cambridge, covering topics from syllogistic reasoning to spatial knowledge representation.